BACTERIAL KIDNEY DISEASE AND ITS EFFECT ON THE … · repeated efforts in helping me to prepare...

BACTERIAL KIDNEY DISEASEAND ITS EFFECT ON THE SALMONID IMMUNE RESPONSE

By

Christine L. Densmore

Dissertation submitted to the faculty of theVirginia Polytechnic Institute and State University

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

in

Veterinary Medical Sciences(Aquatic Animal Medicine)

Approved (Graduate Advisory Committee):

Dr. Stephen Smith Dr. Larisa Ford(Committee chair)

Dr. John Robertson Dr. Rocco Cipriano

Dr. Hugo Veit Dr. Cliff Starliper

April, 1997Blacksburg, Virginia

Key words: bacterial kidney disease, salmonid,immune response, immunity, Renibacterium salmoninarum

BACTERIAL KIDNEY DISEASE AND ITS EFFECT ON THESALMONID IMMUNE RESPONSE

byChristine L. Densmore, D.V.M.

Stephen A. Smith, D.V.M.,Ph.D. - Committee Chair Department of Biomedical Sciences and Pathobiology

Virginia-Maryland Regional College of Veterinary Medicine

Renibacterium salmoninarum , the etiological agent of bacterialkidney disease (BKD) of salmonid fish, is a pathogen of greatconcern among fisheries and the aquaculture industry worldwide.Previous investigations have indicated the pathogenesis of BKD iscomplex. It is a chronic, multisystemic, granulomatous diseasewith a number of potential immunomodulatory effects on the host.Given the current limitations for treatment and control of BKD, itis imperative that the pursuit of development of methods ofprevention, namely management strategies and vaccination, becontinued. To do so, the immunology of BKD must be elucidated inorder to better understand and manipulate the associated immuneresponses to our advantage.

This dissertation is composed of four chapters which relate toBKD and the associated immune responses of three species ofsusceptible salmonid fish as follows:

Exogenous stress factors, through stress-inducedimmunosuppression, have been shown to influence BKD development incultured salmonids. Chapter 1 examines the effects of twoenvironmental stressors common to fish culture, overcrowding andoverfeeding, as they affect BKD development and R. salmoninarumantigen prevalence among juvenile chinook salmon ( Oncorhynchustshawytscha ).

Immunomodulatory interaction between pathogen and host in BKDis widely reported and merits further investigation. Particularly,the immunological parameters affected and the role of theextracellular protein (ECP) of R. salmoninarum are of interest.Chapter 2 examines the in vivo immune response of rainbow trout( Oncorhynchus mykiss ) following exposure to the ECP in terms ofboth humoral and cell-mediated immunological parameters, includingthe immune response against another bacterial pathogen. Chapter 3addresses the in vitro effects of the ECP upon specific splenicimmunocyte functions, phagocytosis and respiratory burst activity,in brook trout ( Salvelinus fontinalis ).

-iii-

The immune-complex mediated hypersensitivity reported to occurwith BKD has considerable ramifications for control measuresinvolving immunostimulation via antigen exposure. Furtherinvestigation is warranted to discern the significance andconsistency of immunological hypersensivity in BKD pathogenesis.Chapter 4 examines the renal lesions, including immunopathologicchanges and indications of immune-mediated disease, of brook troutexposed to R. salmoninarum .

-iv-

Acknowledgments

The completion of this dissertation and my program of graduatestudy would certainly not have been possible without thecontributions and support of many individuals. I wish to extend myheartfelt "thanks" to all of the following:

-the members of my graduate advisory committee for yourguidance, support, and encouragement. Drs. Larisa Ford, RoccoCipriano, Cliff Starliper, John Robertson, and Hugo Veit, you eachcontributed uniquely to my graduate program and I sincerelyappreciate all of your efforts. Particularly, I wish to thank Dr.Stephen Smith for his initiative in making my return to the VMRCVMas a graduate student possible and for his diligent service as myacademic advisor throughout the past four years. Dr. Smith, youare an inspiring role model, an unrivaled character, and a goodfriend; I look forward to the possibility of many years ofcollaborative efforts with you and the rest of our cohorts in theaquatic medicine group at the VMRCVM.

-the staff at the National Fish Health Research Laboratory inKearneysville, WV for your support in my research, your friendship,and many memorable lunchtime conversations. Additionally, I wouldlike to thank Dr. Pete Bullock of the Freshwater Institute for hissupport, the staff of the library at the Leetown Science Center forall their help with the bibliographical aspects of this project,and Dr. David Smith of the Leetown Science Center for numerouslessons in statistical analysis.

-the staffs of the various laboratories and facilitiesassociated with the VMRCVM, especially those concerned with thedisciplines of histology, electron microscopy, biochemistry,toxicology, and aquatic medicine, for all their support andtraining in the technical and applied aspects of my graduateprogram. Particular thanks go out to the group at the aquaticmedicine facilities, especially to Dr. Theresa Hrubec for pavingthe way for a doctoral degree in aquatic medicine at the VMRCVM.I also thank the group from the Biomedical Media Unit for theirrepeated efforts in helping me to prepare visual aids forpresentations.

-the numerous instructors of the VMRCVM, Department ofFisheries and Wildlife Sciences, and Department of Biology (amongothers) of Virginia Tech who instructed me via both the academiccurriculum and informal laboratory sessions. In particular, I wishto thank Dr. Stephen Holladay and Dr. Thomas Caceci for introducingme to the methodologies of flow cytometry and electron microscopy.

-v-

-Dr. John Lee, head of the Veterinary Medical SciencesGraduate Program, for his role in making my graduate programpossible, and Linda Price of the Office of Research and GraduateStudies at the VMRCVM for her assistance with many aspects of mygraduate program.

-Dr. Stephen Kaattari for serving as the external examiner forthis dissertation.

-and finally, my family, for their overwhelming support,enthusiasm, and pride in my accomplishments in this graduateprogram and as long as I can remember. I especially wish to thankmy parents, Jack and Linda Densmore, for a lifetime ofencouragement and eccentricity; thanks to both of you, more than Ican express. Thanks also to you, Mark, for being my #1 fan, thelove of my life, and my best friend; you have been there for me inevery way, shape, and form as I worked to complete this project.Lastly, but least only in terms of stature, I thank my son, Casey,for being the newest and greatest motivating force in my life; Icarried you very close to my heart (literally, only a matter ofinches away) as I prepared a large portion of this dissertation.

Thank you, one and all!!!

-vi-

Table of Contents

Title iAbstract iiAcknowledgments ivTable of Contents viList of Figures viiList of Tables ixList of Abbreviations x

1. Foreword 1

2. Literature Review 2A. Historical Perspective 3B. Characterization of Renibacterium salmoninarum 3

1. Taxonomy, Morphology, Physiology,and Artificial Culture 3

2. Antigenicity 4C. BKD Pathogenesis 6

1. Susceptibility 62. Transmission 73. Incubation and Dissemination 84. Pathology 9

D. Diagnosis 11E. Control 12

1. Management/Husbandry 122. Chemoprophylaxis/Chemotherapeutics 133. Immunization 15

F. Immunology 15G. References 18

3. Afterword/Introduction to Chapters 34

4. Chapter 1: Effects of stocking density and feeding rate upon Renibacterium salmoninarum prevalence among chinook salmon. 36

A. Abstract 37B. Introduction 38C. Materials and Methods 39D. Results 42E. Discussion 45F. Tables 49G. Figures 55H. References 61I. Acknowledgments 64

-vii-

5. Chapter 2: In vivo effects of the extracellular protein of Renibacterium salmoninarum upon immunological parameters in juvenile rainbow trout. 65

A. Abstract 66B. Introduction 67C. Materials and Methods 68D. Results 72E. Discussion 73F. Figures 77G. References 91H. Acknowledgments 94I. Appendices (Tables) 95

6. Chapter 3: In vitro effects of the extracellular protein of Renibacterium salmoninarum on phagocyte function in brook trout. 105


7. Chapter 4: BKD-induced renal pathology of brook trout exposed via consumption of infected tissues. 121


8. General Summary 136

9. Vita 139

-viii-

List of Figures

Chapter 1: (charts)Figure 1 - Antigen prevalence 55Figure 2 - Survival analyses 56Figure 3 - Growth rates 57Figure 4 - ELISA and DFAT correlation 58Figure 5 (a and b) - Water quality covariants,

subclinical progeny population 59Figure 6 (a and b) - Water quality covariants,

clinical progeny population 60

Chapter 2: (charts)Figure 1 - Hematocrit and total serum protein 77Figure 2 - Total leukocyte count 79Figure 3 - Small lymphocyte count 81Figure 4 - Large lymphocyte count 84Figure 5 - Polymorphonuclear cell count 85Figure 6 - Monocyte count 86Figure 7 - Microagglutination antibody titer 88Figure 8 - Phagocytosis 89Figure 9 - Survival 90

Chapter 3:Figure 1 - Phagocytosis of microspheres (photo) 116Figure 2 - Cellular fluorescence (chart) 116Figure 3 - Phagocytosis (chart) 116

Chapter 4: (photos)Figure 1 - BKD-induced renal histopathology 132Figure 2 - Tubular necrosis 132Figure 3 - Granuloma formation 132Figure 4 - Tubular necrosis and

basement membrane phagocytosis 132Figure 5 - Fibroepithelial crescent formation 132Figure 6 - Glomerular basement membrane pathology 132

-ix-

List of Tables

Chapter 1:Table 1 - Organizational scheme - experimental groups 49Table 2 - BKD-related mortality 50Table 3 (a and b) - Growth rates 51Table 4 (a and b) - ELISA data summary 52,53Table 5 - DFAT data summary 54

Chapter 2: (appendices)Appendix A - Hematocrit 96Appendix B - Total serum protein 97Appendix C - Microagglutination antibody titer 98Appendix D - Total leukocyte count 99Appendix E - Small lymphocyte count 100Appendix F - Large lymphocyte count 101Appendix G - Polymorphonuclear cell count 102Appendix H - Monocyte count 103Appendix I - Phagocytosis 104

Chapter 3:Table 1 - Immunofluorescence results 114Table 2 - Phagocytosis results 115

Chapter 4:Table 1 - ELISA results 130Table 2 - Summary of histological observations 131

-x-

List of Abbreviations

ANOVA Analysis of VarianceATCC American Type Cell CultureBKD Bacterial kidney disease (of salmonid fish)DCF 2',7'-dichlorofluorosceinDCF-DA dichlorofluoroescin diacetateDCFH 2',7'-dichlorofluorosceneDFAT direct fluorescent antibody testECAM enteric coated antigen microspheresECP extracellular proteinELISA enzyme-linked immunosorbent assayEMEM Eagle's Minimal Essential MediaERM Enteric redmouth diseaseFALS forward angle light scatterFAT fluorescent antibody testFDA U.S. Food and Drug AdministrationFITC fluoroscein isothiocyanateIg + immunoglobulin positiveIg - immunoglobulin negativeINAD Investigational New Animal DrugIP intraperitonealKDM2 Kidney disease media-2 ( R. salmoninarum culture)KDM-C Kidney Disease Media-CharcoalMFAT membrane filtration fluorescent antibody testMSA major soluble antigenMS-222 tricaine methanesulfateOD optical densityp57 57 kilodalton antigen of R. salmoninarumPAS Periodic acid-Schiff (histological stain)PBS phosphate-buffered salinePBS-T20 phosphate-buffered saline with 0.05% Tween 20PCR polymerase chain reactionPMA phorbol-12-myristate 13-acetatePMN polymorphonuclear cellSALS side angle light scatterTUF target unmasking fluid

-1-

Foreword:

The key to discerning the pathogenesis of any infectiousdisease does not lie solely with an understanding of either thehost or the pathogen, but rather within the realm of theirinteraction. It is for this reason that the study of immunology isso pivotal in medical science. It is the purpose of thisdissertation to utilize this immunological approach to describebacterial kidney disease (BKD) of salmonid fish. Thechapters/manuscripts incorporated into this dissertation allattempt to describe some aspect of the interactions between thepathogen Renibacterium salmoninarum and its salmonid host. Morespecifically, the chapters all address, directly and indirectly,some aspect of the salmonid immune response to R. salmoninarum anddiscuss how this response contributes to the pathogenesis of BKD.The goals of doing so are largely twofold. First and foremost, bydefining the immunologically mediated aspects of the host-pathogeninteraction, we will be better equipped to manipulate thisinteraction advantageously. Therefore, the development ofpractical means of disease control involving immunomodulation wouldbe enhanced. Secondly, this dissertation will hopefully contributeto our general knowledge base of comparative vertebrateimmunopathology. By further describing aspects of the salmonidimmune response to R. salmoninarum , we may enhance ourunderstanding of salmonid immunity as it relates to the evolutionof the vertebrate immune system.

-2-

Literature Review

-3-

A. Historical Perspective

Bacterial kidney disease (BKD) is a multisystemic,granulomatous disease of cultured and feral salmonids worldwide.The causative agent is Renibacterium salmoninarum , a Gram-positivediplobacillus. Bacterial kidney disease was first reported oversixty years ago with one of the earliest known outbreaks occurringin feral salmon during the spring/summers of 1930 through 1932 inthe Aberdeenshire Dee and the River Spey of Scotland (Smith, 1964).In 1934, an outbreak attributed to BKD also occurred in culturedtrout at a hatchery in Massachusetts (Belding and Merrill, 1935).Subsequently, other outbreaks among Pacific Northwest salmon stocksand eastern trout stocks in the U.S., as well as among salmonids inEurope, had been associated with the same as yet unclassifiedpathogen. Due to its morphology and physiology, the pathogen wasinitially classified as a Corynebacterium (Ordal and Earp, 1956).

By the latter half of the twentieth century, BKD (also knownas Dee disease, salmonid kidney disease, corynebacterial kidneydisease, white boil disease) was recognized as a significantdisease among a wide variety of salmonid species across NorthAmerica including the U.S.A. and Canada; Europe including theUnited Kingdom, Germany, Scandinavian nations, and Iceland; andAsia including Japan. Research focused on BKD had intensifiedalong with the disease's global dissemination. Various aspects ofBKD including pathogenesis, diagnosis, and control strategies werebeing widely investigated. Additionally, the taxonomy of thecausative agent was also becoming more precisely defined. In 1980,the pathogen was reclassified to its own genus and namedRenibacterium salmoninarum (Sanders and Fryer, 1980).

B. Characterization of Renibacterium salmoninarum

1. Taxonomy, Morphology, Physiology, and Artificial CultureTaxonomic reclassification of the pathogen as Renibacterium

salmoninarum was based on the organism's biochemical composition.Specifically, the peptidoglycan layer, cell wall polysaccharides,and the guanine + cytosine content all differed significantlyenough from Corynebacterium and other established genera to warrantcreation of the new genus Renibacterium (Sanders and Fryer, 1980).Stackebrandt et al. (1988) further described Renibacteriumtaxonomically as belonging to the Actinomycetes subdivision of theCoryneform group of bacteria; this study also revealed closesimilarities between the genus Renibacterium and the generaArthrobacter and Micrococcus . Renibacterium salmoninarum is theonly species thus far described in this genus.

-4-

Morphologically, R. salmoninarum is characterized as small,measuring approximately 0.4 x 0.8 µm (Bullock and Herman, 1988).The organism is strongly Gram-positive, non-motile, non-encapsulated, and non-spore forming. The shape is slightlypleomorphic. In artificial culture the bacteria tends toconsistently appear as a diplobacillus (Fryer and Sanders, 1981),while cells obtained from host tissues are more prone toirregularities in shape, including clubbing (Young and Chapman,1978). Distinctive features of R. salmoninarum 's biochemicalcomposition, include a 53% guanine + cytosine content,peptidoglycan amino acids including lysine, alanine, glycine, andglutamic acid and a unique cell wall polysaccharide compositionincluding galactose, rhamnose, N-acetyl glucosamine, and N-acetylfucosamine (Fryer and Sanders, 1981; Fiedler and Draxl, 1986).

Physiologically, R. salmoninarum is described as fastidiousand slow-growing, proteolytic, catalase-positive, and non-liquefactive for gelatin. Renibacterium salmoninarum ishydrophobic; this property represents a potential virulence factorin that it may increase cellular adhesion and uptake by host targetcells (Daly and Stevenson, 1987). The organism has specifictemperature requirements, growing best at 15 C, more slowly at 22 C,o o

and not at all at 37 C. Nutritional requirements for growth areo

also limiting for artificial culture of the bacterium.

Culture of the organism was first achieved through the use ofnutrient enhanced agar medium containing various animal tissuecomponents (Earp et al., 1953). This type of medium containingserum or meat infusion generally yields appreciable growth of R.salmoninarum in approximately two weeks. Modification of themedium with 0.05 to 1% cysteine decreases this time span to 7-10days (Fryer and Sanders, 1981). A further modification of acysteine blood agar medium by Evelyn (1977) using the addition ofpeptone and the elimination of sodium chloride, tryptose, and beefextract is named Kidney Disease Medium-2 (KDM2) and furtherenhances the growth of R. salmoninarum . An additional modificationresulted from a satellitism growth technique that evolved intomedium supplementation with a metabolite product from spent R.salmoninarum cultures (Evelyn et al., 1990; Teska, 1994).Currently, the various modifications of KDM2 medium appear to bethe most widely used medium for artificial growth of R.salmoninarum . Among these modifications are a selective KDM2medium (Austin et al., 1983) and a charcoal based medium, KDM-C(Daly and Stevenson, 1985). 2. Antigenicity

The antigenicity of R. salmoninarum is described in terms ofits relevance to both diagnostic parameters and diseasepathogenesis. Various cell surface antigens and antigens

-5-

associated extracellularly with R. salmoninarum (i.e. free proteinproduced by R. salmoninarum found in the spent supernatant of brothcultures) have been identified. The most common antigen found onthe cell surface and as an extracellular component is a 57kilodalton protein (p57), also known as the major soluble antigen(MSA) or Antigen F (Getchell et al., 1985). Debreuil et al. (1990)identified this p57 antigen to be a cell surface fimbriae less than2 nm in diameter. Various other antigens of higher and lowermolecular weights are identified (Fiedler and Draxl, 1986; Getchellet al., 1985), and many are thought to be potential breakdownproducts of the p57 antigen (Bandin et al., 1992). The p57 antigenis generally regarded as unstable, with degradation occurring inassociation with elevated temperature and time. Griffiths andLynch (1991) attributed this instability to autolysis. Still,other findings indicated that the p57 antigen is a heat stable protein (Getchell et al., 1985).

Renibacterium salmoninarum antigens, and particularly the p57antigen, have been instrumental to the evolution of diagnosticmethods for detection of BKD. Monoclonal antibodies developedagainst R. salmoninarum for serological tests recognize variousepitopes of the p57 antigen (Weins and Kaattari, 1989) and haveproven extremely useful in identifying the bacterium (Part D -Diagnostics) and characterizing its effects (Weins and Kaattari,1991).

A number of physiological features of R. salmoninarum ,including various virulence factors have been associated with itscell surface and/or extracellular protein (ECP). Autoaggregationof R. salmoninarum organisms and the agglutination of heterologouscell types, including rabbit erythrocytes and salmonid spermatozoa,have been attributed to the p57 antigen on the cell surface (Dalyand Stevenson, 1989). Senson and Stevenson (1994) also related theautoagglutination ability of R. salmoninarum to the presence ofcell surface p57. A variety of immunomodulatory effects associatedwith R. salmoninarum in BKD pathogenesis have also been related tothe bacterial proteins. In keeping with the association of p57with these various virulence factors, there is evidence that theoverall virulence of R. salmoninarum is related to the presence ofthis antigen. Bruno (1990) identified the presence of p57 inextracts of virulent R. salmoninarum cells and its absence insimilar extracts of avirulent cells.

Other recent research studied the structure of the p57 antigenthrough amino acid or codon sequencing techniques . Chien et al.(1992) described the complete sequence coding for this antigen,including indications of amino acid residues present in theprecursor and mature proteins (557 amino acids, MW=57,190 D).Radacovici and Debreuil (1991) utilized enzymatic digestion

-6-

techniques and amino acid sequencing to evaluate the structure andantigenicity of fragments of the p57 protein.

C. BKD Pathogenesis

1. SusceptibilityA number of factors have significant bearing upon a salmonid

host fish's susceptibility to BKD. This disease appears to beexclusively a disease of salmonids, as only salmonid fish,including Atlantic salmon ( Salmo salar ), a variety of species ofPacific salmon ( Oncorhynchus spp.), trout ( Salmo spp., Salvelinusspp.) and grayling ( Thymallus ) have been reported to be infectedwith R. salmoninarum and develop BKD (Kettler et al., 1986;Bullock and Herman, 1988). Attempts to challenge non-salmonid fishwith R. salmoninarum are not well represented in the literature,although one such attempted challenge in the Pacific lamprey( Lampetra tridentata ) did not show infection (Bell and Traxler,1986). Experimental BKD infection by means of intraperitonealinjection was achieved with sablefish ( Anoplopoma fimbria ),although no naturally occurring infections are reported in thisspecies (Bell et al., 1990).

Among the salmonid species, susceptibility to BKD differssignificantly, with the Pacific salmon species being the mostsusceptible and rainbow trout ( Oncorhynchus mykiss ) the least(Evelyn et al., 1988 ,Bruno, 1988; Sakai et al., 1991). Geneticvariability within a species may also yield intraspecificdifferences in susceptibility. For instance, Winter et al. (1980)reported a differential resistance to BKD among stocks of cohosalmon ( Oncorhynchus kisutch ) and steelhead trout ( Salmo gairdneri )based upon genotype. Heritability estimates of BKD susceptibilityamong chinook salmon ( Oncorhynchus tshawytscha ) and Atlantic salmon( Salmo salar ) indicated that the heritability component is greaterfor BKD than for some other bacterial diseases of these salmonidspecies (Beacham and Evelyn, 1992; Gjedrem and Gjoen, 1995).Genetic manipulation of stocks for BKD resistance, while promising,has not been frequently reported in the literature. There is noevidence of enhancement of BKD resistance through geneticmanipulations of ploidy; Bruno and Johnstone (1990) found nodifferences in susceptibility between triploid Atlantic salmon andtheir diploid counterparts.

Age of fish appears to have some affect on BKD susceptibility,although juvenile and adult salmonids of all stages are vulnerable.Stress susceptibility and adequacy of host immunological capabilityare influenced by age and have a direct bearing upon subsequent BKDdevelopment. For instance, smoltification imposes a physiologicalstress which increases BKD susceptibility (Paterson et al., 1981b).

-7-

Bacterial kidney disease is largely regarded as a problem ofcultured (hatchery) fish, although feral populations appear to bequite susceptible as well. Salmonid culture facilities haveexperienced losses of up to 80% of Pacific salmon stocks in BritishColumbia and up to 40% of Atlantic salmon stocks in Scotland;recurring BKD-related losses of approximately 10% of rainbow troutstocks per year were common for Scottish fish farms in the 1980's(Bruno, 1988). Renibacterium salmoninarum has been detected inferal salmonid populations worldwide, and BKD epizootics arelikewise reported in feral fish (Mitchum et al., 1979).

A variety of extrinsic, i.e. environmental, factors alsosignificantly influence a salmonid's susceptibility to BKD.Temperature is an environmental parameter which impacts upon BKDdevelopment, as it affects both the virulence of the pathogen andthe immunological response of the host. Wolf and Dunbar (1959)described an increased virulence of R. salmoninarum in brook trout( Salvelinus fontinalis ) and rainbow trout at 7 C in contrast too

12.5 C. Sanders et al. (1978) found that BKD-related mortalityo

decreased among coho salmon and steelhead trout as watertemperatures increased above 12.2 C. More recently, Ortega eto

al.(1994) reported a correlation between decreasing watertemperature and decreased mortality from BKD among rainbow trout.Similarly, BKD development and occurrence of epizootics varyseasonally, presumably related to fluctuations in water temperature(Belding and Merrill, 1935; Fryer and Sanders, 1981). Other waterquality parameters including salinity are reported to affect BKDdevelopment. Significant levels of BKD mortality are noted to occuras anadromous salmonid smolts migrate from fresh to salt water(Banner et al.,1983; Banner et al.,1986; Paterson et al., 1981).

Nutritional parameters represent another extrinsic variableinfluencing BKD susceptibility among salmonids. Dietarysupplementation with various trace minerals and vitamins isreported to substantially reduce prevalence of BKD in salmonidfish. Increased levels of iodine and fluorine have been associatedwith decreased BKD among rainbow trout and Atlantic salmon (Lall etal., 1985; Bowser et al., 1988). Possible correlations betweenother trace dietary constituents including iron, cobalt, copper,and manganese have been observed (Paterson et al.1981), andsurvival of sockeye salmon ( Oncorhynchus nerka ) with BKD was foundto be inversely correlated with Vitamin C levels in the diet (Bellet al., 1984). In contrast, no relationship is evident betweendietary levels of either Vitamin E or selenium and R. salmoninarumprevalence in chinook salmon (Thorarinsson et al., 1994).

2. TransmissionRenibacterium salmoninarum is transmitted directly, fish to

fish, both horizontally and vertically. Horizontal transmissionoccurs through direct contact between infected, both diseased and

-8-

carrier, and uninfected individuals as well as through theconsumption of infected tissue (Mitchum and Sherman, 1981; Wood andWallis, 1955). Fecal-oral transmission also occurs as R.salmoninarum is excreted in the feces of infected host fish and mayremain a viable inhabitant of organic waste in the water for up to21 days (Austin and Rayment, 1985; Balfry, 1990). Reservoirs of R.salmoninarum for horizontal transmission may be clinically diseasedindividuals or subclinically infected carriers. Horizontaltransmission via water-borne organisms in the absence of fish tofish contact has not been demonstrated and is thought to beunlikely (Kaattari et al., 1989). However, vertical transmissionof R. salmoninarum, involving intra-ovum infection of the egg andsubsequently the developing progeny, seems to be the moresignificant mode of naturally occurring infection (Evelyn et al.,1986a; Evelyn et al., 1984).

No specific vectors or non-salmonid reservoirs have as yetbeen identified in association with R. salmoninarum . The bluemussel, Mytilus edulis , is reported to clear seawater of R.salmoninarum , yet this mollusk does not appear to maintain viableorganisms as a reservoir source(Paclibare et al., 1994). However,poor husbandry or sampling techniques used with salmonids mayprovide a source of mechanical vectors enhancing horizontaltransmission of R. salmoninarum . For instance, Elliott et al.(1994) showed that coded-wire tagging of salmon may transmit thepathogen via contaminated needles.

3. Incubation and Dissemination Time from initial exposure to fulminate clinical disease is

variable (up to several months) depending on numerous conditions.Incubation varies with the virulence of the pathogen and theimmunocompetence of the salmonid host, both of which may beinfluenced by environmental variables (water temperature, salinityetc.) as previously described. Although BKD is generally regardedas a chronic disease, acute epizootics are not uncommon, and aprecise incubation period has not been defined.

Initial entry of the R. salmoninarum pathogen into the host isknown to occur through the skin or mucous membranes such as the eyeor gastrointestinal tract. The means of subsequent disseminationof the pathogen has not been determined, although it is likely thatthe pathogen travels through the host either intracellularly inphagocytes or extracelluarly within the circulatory system andother body fluids (Evendon et al., 1993). Whichever the case,rapid systemic dissemination following exposure is likely. Bruno(1986a) showed R. salmoninarum intracellularly in the kidney andspleen of host fish less than 45 minutes after exposure viaintraperitoneal injection.

-9-

4. PathologyPathological changes in host tissue resulting from BKD

infection are potentially quite severe and multisystemic. Whileinflammatory processes are undoubtedly responsible for thesevarious lesions, several theories exist regarding the pathogenesisof lesions. Various explanations include: bacterial cytotoxinsand/or proteolytic factors which are released from the pathogendamage host tissues; mechanical damage to host cells is caused bythe bacteria, triggering the extracellular release of cell contentsincluding enzymes and inflammatory mediators; and immunologicalhypersensitivity reactions, in which the host immune system attacksautologous tissue, are initiated by R. salmoninarum antigen(Evendon et al., 1993; Young and Chapman, 1978; Sami et al., 1992).It is quite possible that some combination of these scenariosoccurs. The relative contribution from each mechanism maytherefore partially explain the variability encountered in theclinical and pathological manifestations of BKD.

Gross external lesions in R. salmoninarum -infected fish areoften variable, but may include some combination of the following:petecchial hemorrhages, particularly around the base of fins orexternal orifices; unilateral or bilateral exophthalmia; skin blebsor blisters (filled with bacteria and serous, serosanguinous, orpurulent fluid) which may eventually rupture and/or coalesce toform large ulcerative lesions; skin abscesses; altered skinpigmentation; and abdominal distension due to free fluidaccumulation in the body cavity.

Gross internal lesions may likewise be varied, but oftenconsist of: diffuse swelling of internal organs, particularly theliver, spleen, and kidney; focal abscesses and/or diffuse purulentexudate from internal organs, particularly the liver, spleen, andkidney; petecchial hemorrhage on internal organs and the muscle ofthe body wall; discoloration of internal organs, usuallymanifesting as a pale, mottled liver, reddened spleen, and grey towhite kidney; pseudomembranous covering of various internalorgans; and abundant serous/serosanguinous fluid accumulation inthe body cavity (Evendon et al., 1993; Bullock and Herman, 1988;Fryer and Sanders, 1981).

Histologically, BKD pathology also shows considerablevariability, but is generally associated with signs of chronicinflammation. The hallmark of BKD histopathology is multisystemic,diffuse granulomatous inflammatory lesions (Fryer and Sanders,1981). Granuloma formation may occur in various tissues and isaccompanied or preceded by cellular tissue infiltration bymacrophages, neutrophils, and thrombocytes. Also evident aretissue edema and degeneration of tissue structures such as renaltubules or glomeruli, culminating in multifocal necrotic centers.

-10-

The organs most commonly and severely affected include the liver,spleen, excretory kidney, and hematopoietic kidney. Additionally,the heart, gonad, gill, swimbladder, gastrointestinal tractincluding the pancreas, and central nervous system includingsensory organs are frequently reported to show lesions induced byBKD. Extensive necrotizing myocarditis, mild to severemeningoencephalitis/ventriculitis, and similar inflammatory changesare described in these other organ systems as well (Bruno, 1986a;Speare et al., 1993). In all affected tissue, numerous R.salmoninarum organisms are frequently encountered bothintracellularly and extracelluarly. An immunological mechanism forinduction of renal lesions has been described in the literature(Section F-Immunology).

Ultrastructural lesions are consistent with the cellulardamage commonly associated with severe acute to chronicinflammation. Cellular changes in BKD-induced lesions includeswelling and rupture of mitochondria, dilation of endoplasmicreticulum, fatty infiltration of cytoplasm, lysosomal and nuclearmorphological alterations, vacuolization, collagen deposition alongbasal borders, hyaline granule accumulation (renal tubule cells),and eventual disruption of cellular membrane integrity (Young andChapman, 1978).

The morphological evidence of disease in BKD is associatedwith a variety of physiological alterations. BKD is associatedwith changes in hematology parameters including decreasedhematocrit, decreased hemoglobin, decreased red cell diameter, anddecreased ratio of mature to immature erythrocytes (Bruno andMunro, 1986; Suzumoto et al., 1977). This BKD-induced anemia ispresumably due to erythrocyte retention in the spleen.Additionally, a leukocytosis characterized by monocytosis andneutrophilia as well as a thrombocytosis, were noted in associationwith BKD (Bruno and Munro, 1986). Changes in serum chemistryparameters are also reported in salmonids with BKD. These changesinclude hypoproteinemia, hypocholesterolemia, hyponatremia,hyperkalemia, hypoglycemia, and hyperbilirubinemia (Bruno, 1986b;Fryer and Sanders, 1981). The decreased serum protein levels arepresumed to be related to either renal or hepatic impairment (i.e.excessive protein excretion or diminished protein synthesis). Thehyperbilirubinemia is likely attributable to liver dysfunction.The electrolyte imbalance may be related to either extensivemultisystemic cellular damage and subsequent disruption of cellularcross-membrane electrolyte gradients or to primary renal failure.In any event, these imbalances contribute significantly toalterations in intra/extravascular hydrostatic pressures and fluidbalances, resulting in further physiological dysfunction includingfree fluid accumulation in extravascular spaces and circulatorycompromise (Bruno, 1986b). Severe fluid imbalance and circulatory

-11-

failure, as well as vital organ dysfunction (primarily renal), arepostulated to be the most significant causes of death from BKD(Hayakawa et al., 1989; Evendon et al., 1993).

D. Diagnosis

The detection of R. salmoninarum and diagnosis of BKD havebeen greatly refined over the past few decades. Historically,early diagnosis of BKD was accomplished through correlatingobserved clinical signs with positive identification of Gram-positive diplobacillary organisms in affected tissues.Subsequently, refined culture techniques allowed for bacterialisolation in artificial media to help confirm the presence of R.salmoninarum . These diagnostic methods, however, were oftenimpractical and inconclusive, as Gram-staining does not precludeother coryneform bacteria and successful isolation of R.salmoninarum in culture medium may take over three weeks under thebest of circumstances (Fryer and Sanders, 1981). Therefore, eventhough bacterial isolation by culture has proven quite reliablediagnostically, the time requirement makes this an undesirableoption for diagnosis when used alone.

A variety of serological techniques have greatly improved theease and reliability of BKD diagnosis. Chen et al. (1974)developed an immunodiffusion test to detection soluble antigen ofR. salmoninarum in infected host tissue. Similarly, Kimura (1978)described a Staphylococcus coagglutination technique for detectingR. salmoninarum antigen in tissues. Bullock and Stuckey (1975)described a fluorescent antibody technique (FAT) developed againstthe cell surface antigen of R. salmoninarum . This test has provento be a significant advancement in BKD diagnostic capability as itimproved upon speed and sensitivity (Fryer and Sanders, 1981). Anumber of variations upon this indirect FAT have proven quiteuseful in clinical and research related diagnosis. These includethe direct FAT (Bullock et al., 1980), a quantitative FAT(Cvitanich, 1994), a membrane filtration FAT (Elliott and Barila,1987), and a technique for immunofluorescent detection of R.salmoninarum antigen in fixed, paraffin-embedded host tissue(Evenson, et al., 1994).

Pascho and Mulcahy (1987) made another formidable contributionto BKD diagnostics with the development of the enzyme-linkedimmunosorbent assay (ELISA) for detection of R. salmoninarumantigen. Further development of this test procedure, including theuse of a monoclonal anti- R. salmoninarum antibody (Hsu et al.,1991) and the dot-blot ELISA (Sakai et al., 1987) increased itsspecificity, sensitivity, and practical application. Still other

-12-

serological techniques including counterimmunoelectrophoresis(Cipriano et al., 1985), immunohistochemistry of melanin-containingcells (Jansson et al., 1991), Western blot, and theimmunoelectrotransfer blot (Olivier et al., 1992) are also used todetect R. salmoninarum antigen in infected host tissues.Comparatively, the serological techniques which identify solubleantigen of R. salmoninarum appear to be more sensitive than thevarious FAT techniques (Griffiths et al., 1991; Cipriano et al.,1985), although both of these diagnostic options are far morereliable than diagnoses based upon clinical signs, histopathology,and/or Gram-staining, particularly in asymptomatic fish. The useof monoclonal antiserum, as opposed to polyvalent anti- R.salmoninarum serum, also improves serological sensitivity andspecificity. However, cross-reactivity with other micro-organismsis not entirely eliminated, even with the monoclonal preparations(Bandin et al., 1993c). In light of the complementary strengthsand shortcomings of the various diagnostic techniques available forBKD, perhaps the most desirable approach would involve multipletechniques incorporating serology, bacterial isolation, andhistology (Teska et al., 1995; White et al., 1995).

Other promising approaches to BKD diagnostics involve the useof genetic probes and gene amplification techniques. Both DNA andRNA probes have been developed for R. salmoninarum and are beingused successfully to identify bacteria in host tissues (Leon etal., 1994a; Magnusson et al., 1994). Gene amplification, namelythe polymerase chain reaction (PCR), is widely used in conjunctionwith genetic probes to further enhance test sensitivity (Leon etal., 1994b; Brown et al., 1994). Genetic probe diagnosis of BKDhas the advantages of increased sensitivity and specificity overserological methodologies without sacrificing the component oftime, as occurs with bacterial isolation techniques (Hariharan etal., 1996; Butendieck et al., 1995).

E. Control

1. Management/HusbandryControl measures for BKD include a number of approaches.

Various factors associated with management technique and animalhusbandry have proven to be effective inhibitors of R. salmoninaruminfection and disease development. Environmental parameters whichact as risk factors in BKD pathogenesis, i.e. water temperature andwater chemistry, may be advantageously manipulated in fish culturesituations to decrease both the impact of nonspecific environmentalstress and the pathogen's virulence. Nutrition may also serve asa means of BKD control as dietary modifications, particularly withregard to the trace mineral and vitamin components previouslydescribed, may diminish susceptibility of salmonids to BKD.

-13-

Likewise, genetic manipulation of stocks may be undertaken todecrease overall host susceptibility to BKD.

More specifically, impairing transmission of R. salmoninarumfrom fish to fish is an effective means of disease control.Historically, a vast reduction of the level of horizontaltransmission was accomplished in the 1960's with the advent ofpasteurization of fish feeds containing R. salmoninarum -infectedsalmonid tissue (Fryer and Sanders, 1981). Horizontal transmissionamong live fish may likewise be reduced through the utilization ofwater ozonization or chlorine-based disinfectants in hatcheries.(Austin, 1983; Pascho et al., 1995), as would minimizing the levelof organic debris in the water of culture facilities (Austin andRayment, 1985).

Reduction of vertical transmission by husbandry-relatedmodifications have also been shown to be effective means of BKDcontrol. Brood stock segregation based on R. salmoninarum antigenprevalence among adult salmonids has been shown to dramaticallyreduce the levels of antigen prevalence and BKD-related mortalityamong progeny (Pascho et al., 1991; Elliott et al., 1995).Similarly, Aedo and Bustos (1991) described a technique for broodstock segregation and aseptic spawning effective for obtaining R.salmoninarum -free fertilized eggs. Disinfection techniquescommonly employed in spawning do not, however, appear to be asuitable control of vertical transmission (Evelyn et al., 1984). 2. Chemoprophylaxis/Chemotherapeutics

The use of antibiotics as preventative or treatment for BKDhas met with the most success of any control measure. In the early1950's, sulfonamide antibiotics incorporated into the diet wereused to control BKD, but long term elimination of R. salmoninarumfrom the salmonid population could not be achieved with thesebacteriostatic therapeutants (Elliot et al., 1989). Wolf andDunbar (1959) examined the efficacy of ten antibiotics in treatingBKD, and found the most satisfactory therapeutic regime waserythromycin at 100 mg/kg/day for 21 days. Austin (1985) similarlyevaluated over 70 antibiotics for efficacy against BKD and found afew (clindamycin, kitasamycin, penicillin G, spiramycin, anderythromycin) to be effective in the treatment of clinical BKD inthe earlier stages. Additionally, cephradine, lincomycin, andrifampicin were found to be satisfactory chemoprophylactic agents.This study also reported that a 10 day treatment regime witherythromycin was satisfactory, as compared to the previouslyadvocated 21 day regime, noting that decreasing the environmentalleaching of antimicrobials wherever possible would decrease therisk of selecting for antibiotic-resistant microflora. Morerecently, enrofloxacin has been shown to be an effectivechemotherapeutant against BKD (Hsu et al., 1994). Overall,

-14-

erythromycin is widely regarded as the antimicrobial of choice inthe control of BKD.

Erythromycin chemotherapy/chemoprophylaxis poses a number ofimportant considerations. The route of administration is a crucialvariable. The efficacy of oral administration of erythromycin isquestionable, because of the low palatability in diets (Moffitt,1992). Limited study has been devoted to immersion therapy, andthis route of administration has not thus far been provensatisfactory (Elliott et al., 1989). Parenteral injection wasshown to be efficacious in numerous studies, and appears to be themost widely used and advocated route of administration(Elliott etal., 1989; Evelyn et al., 1986b; Armstrong et al., 1989; Lee andEvelyn, 1994; Brown et al., 1990).

The age of the salmonid during erythromycin administration iscrucial to treatment efficacy. Treatment of salmonid eggs witherythromycin prior to water-hardening was shown to be ineffectivein eliminating R. salmoninarum from progeny (Evelyn et al., 1986a;Bruno and Munro, 1986). Parenteral erythromycin injection of adultbrood stock prior to spawning, however, appears to be quiteeffective in the prevention of vertical transmission of R.salmoninarum (Evelyn et al., 1986b; Armstrong et al., 1989; Brownet al., 1990; Lee and Evelyn, 1994). The pre-spawning timing oftherapy is also important, with the last parenteral injection given34 to 56 days pre-spawning being the most efficacious (Evelyn etal., 1986b).

Antibiotic, particularly erythromycin, chemotherapy in thecontrol of BKD is not without its disadvantages. There is thepotential of creating iatrogenic lesions with erythromycin use insalmonids, including renal tubular lesions and lysosomal congestiveoverloading. These histological changes induced by therapeuticlevels of erythromycin have, thus far, been shown to be largelyreversible (Hicks and Geraci, 1984). The non-approved status ofmost chemotherapeutants, including erythromycin, in fish culture isalso restrictive, although there is the possibility for INADapproval by the FDA when applicable (Moffitt, 1991). FDArestrictions imposed on the use of erythromycin, along with othertherapeutic agents, is, however, certainly warranted, as concernsover the development of antibiotic resistant microbes are quiterelevant to fish culture. Bell et al. (1988) already reported thedevelopment of an erythromycin-resistant strain of R. salmoninarum ,a serious finding given the limited availability of currentalternatives for BKD control.

-15-

3. ImmunizationProphylactic control of BKD through immunization has been the

focus of considerable research effort, especially over the pastdecade. Passive immunity against BKD with serum from BKD-challenged survivors was shown to diminish BKD-related mortality(Campbell et al., 1994), but this concept is not feasible for largescale prophylactic application. Active immunization, primarilythrough the stimulation of humoral immune responses, has beenattempted on a larger scale, but produced much conflicting data.Preparations include bacterins (heat killed, pH lysed, formalininactivated) with and without adjuvants incorporated intomonovalent or multivalent vaccines (Elliott et al., 1989). Variousroutes of administration utilized include direct immersion,hyperosmotic infiltration, and parenteral (intraperitoneal)injection (McCarthy et al., 1984; Paterson et al., 1981a), but nonehave proven conclusively to be beneficial. In one study (Sakai etal., 1989), intraperitoneal injection of a bacterin produced anagglutinating antibody titer and increased phagocytic activity ofimmunocytes, but subsequent experimental challenge demonstrated alack of protective immunity. However, oral administration of a BKDpreparation via enteric coated antigen microspheres (ECAMs) wasassociated with decreased levels of R. salmoninarum protein,indicating potential efficacity (Piganelli, 1995).

F. Immunology

The immunological reaction of a host salmonid to R.salmoninarum is integral to BKD pathogenesis. A unique feature ofBKD which is pertinent to its complex immunopathology is thepotential immunomodulatory effect of R. salmoninarum upon the host.Immunosuppression associated with BKD is described by a number ofinvestigators. The leukoagglutination of salmonid pronephric,splenic, and circulating leukocytes and decreased antibodyresponses related to exposure to the p57 antigen in vitro arereported by Kaattari et al. (1989). Iwama and Greer (1980) showedthat susceptibility to the toxic effects of a non-infectious agent,sodium pentachlorophenate, increased in juvenile coho salmonfollowing exposure to R. salmoninarum . Alternatively, R.salmoninarum may act as an immunostimulant. Amend and Johnson(1984) suggested that R. salmoninarum potentiates humoralprotection against Aeromonas salmonicida when administered as abivalent bacterin.

Salmonid defenses against R. salmoninarum involving non-specific immunity have not been widely investigated. Neutrophiliais reported in association with BKD (Bruno and Munro, 1986), butneutrophil-mediated effects are not described in any detail.

-16-

Literature regarding non-specific acellular defenses such ascomplement and acute-phase proteins is also sparse. Rose andLevine (1992) showed a depletion of human and piscine complement inthe presence of R. salmoninarum , suggesting that the alternativecomplement pathway and bacterial opsonization are involved in BKDpathogenesis. Additionally, the presence of complement in vitrois associated with increased intracellular survival of phagocytosedR. salmoninarum (Bandin et al., 1995). Decreased serum proteinlevels are also associated with clinical BKD (Bruno, 1986).

As previously alluded to (Part E-Control/Immunization),thehumoral immune response in BKD is described in a number ofinvestigations, yet it is questionable as to whether this responseis truly protective. McCarthy et al. (1984) evaluated a variety ofbacterins (formalin and pH inactivated at different concentrations)and routes of vaccination (hyperosmotic, immersion, andintraperitoneal injection) to determine efficacy against challengeinfection; they concluded that the resulting humoral immunity wasdependent upon the type of bacterin and route of administration;protective immunity was demonstrated with intraperitoneal injectionof formalin inactivated R. salmoninarum bacterins, but not with theother routes. Paterson et al. (1981a) also found intraperitonealinjection to be far more efficacious a vaccine route thanhyperosmotic immersion. This study also concluded that while thetwo year secondary immune response to R. salmoninarum in Atlanticsalmon vaccinated as pre-yearling parr was poor, post-yearling parrvaccinated with R. salmoninarum cells and Freund’s completeadjuvant (FCA) had decreased BKD-related lesions compared tocontrols. Alternatively. Sakai et al. (1993) identified a humoralresponse to R. salmoninarum in rainbow trout following vaccinationwith various bacterin preparations, but this response was not foundto be significantly protective under any conditions. Additionally,Bruno (1987) found a lack of correlation between the level ofagglutination titer and level of R. salmoninarum infection inAtlantic salmon. Other investigators (Lovely et al., 1994;CaboBravo, 1993) found that asymptomatically infected fish may havelow to absent antibody titers. Collectively, investigationsdescribing humoral immunity in BKD suggest that the humoralresponse to R. salmoninarum infections in salmonids does notconsistently, if at all, provide protective immunity, and, in lowlevel infective states, may not produce detectable antibody titers.Wood and Kaattari (1996) described an enhanced humoral response inchinook salmon, directed against the carbohydrate moieties of R.salmoninarum cells, associated with removal of the p57 antigen.This increased immunogenicity of R. salmoninarum without p57 lendsfurther support to the proposed immunomodulatory nature of thisantigen.

-17-

In light of these developments regarding the role of humoralimmunity, the cell-mediated immune response is emerging as asignificant factor in BKD pathogenesis and, ultimately, as a focusin BKD control. Renibacterium salmoninarum may be likened to otherpathogenic bacteria such as Mycobacterium , Listeria , and Legionellabased upon their ability to thrive intracellularly in host tissue.Immunologically, this intracellular existence is quite significantin BKD as it provides a relatively protected environment for thepathogen from the host’s defenses. In addition, the intracellularlocation of the pathogen indicates a more prominent role of cell-mediated immunity in the disease pathogenesis (Kaattari et al.,1989). In particular, R. salmoninarum appears to have a tendencyto locate intracellularly in salmonid phagocytes, namelymacrophages (Young and Chapman, 1978). Not only does R.salmoninarum possess the ability to avoid host killing mechanismsand thereby survive intracellularly in salmonid phagocytes (Bandinet al., 1993a), but many investigators note the likelihood thatthe pathogen also multiplies intracellularly (Bandin et al., 1993b;Klein, 1992). Logistics of the intracellular existence of R.salmoninarum are not fully elucidated. It is theorized that thevirulence factors associated with the p57 cell surface protein maycontribute as the p57 protein may serve as a fimbral adhesionmolecule (Dubreuil et al., 1990). The means by which R.salmoninarum gains entry into the cell are also undetermined;postulated mechanisms include endocytosis, Fc receptor binding(Evendon et al., 1993), and C3b receptor binding (Rose and Levine,1992).

It is also postulated that R. salmoninarum directly interfereswith the host’s killing mechanisms, especially phagocyticrespiratory burst activity, in order to avoid the host’s immunedefenses (Bandin et al., 1993b). Interestingly, this effect ofdecreased respiratory burst activity in the presence of R.salmoninarum appears to be related to a fish’s susceptibility to R.salmoninarum infection, probably reflecting the decreased capacityof the host immune system to kill the bacteria (Sakai et al.,1996). Conversely, Hardie et al. (1996) showed that macrophagestimulation with macrophage activating factor(MAF) could reversethe depleted respiratory burst activity and inhibit R. salmoninarumgrowth following in vitro exposure to the pathogen.

Histologically, the affinity of R. salmoninarum for phagocyticcells and the importance of cell-mediated immunity in BKD isevidenced through the localization of R. salmoninarum organisms inhematopoietic tissues in association with melanomacrophage centersand sinusoidal-lining cells (Flano et al., 1996). Other typicalpathological findings associated with BKD reflect a chronic,granulomatous and necrotizing disorder, further emphasizing the

-18-

importance of cell-mediated defenses (Evendon et al., 1993; Bruno,1986). Investigations in vaccine development provide additionalevidence supporting the importance of cell-mediated immunity in BKDpathogenesis, as the most efficacious experimental vaccinepreparations have incorporated FCA, an immunostimulant whichtargets the cell-mediated immune response (Kaattari et al., 1989).

Immunological hypersensivity, the exaggerated response to aforeign agent resulting in autologous tissue damage, is anotherprominent feature of BKD pathogenesis. Young and Chapman (1978)described ultrastructural changes of the renal glomeruli includingthe appearance of flocculent deposits associated with theglomerular basement membrane and the mesangium, mesangialproliferation, thickening and irregularity of the basementmembrane, and extensive inflammatory cell invasion. Theirinvestigation noted the pathological similarities between BKD andmammalian glomerulonephritis and nephrotic syndrome which arecharacteristic of a Type III (immune complex deposition)hypersensitivity. Sami et al. (1992) described a chronicmembranous glomerulonephritis in experimentally R. salmoninarum -infected rainbow trout with similar glomerulopathy and alsoidentified immunofluorescence along the glomerular basementmembrane, attributing it to immune complex deposition. Theseobservations are supported by the work of Kaattari et al. (1989)which identified soluble immune complexes in the serum of cohosalmon experimentally infected with R. salmoninarum .

The possibility of a Type IV (delayed, cell-mediated)hypersensitivity associated with BKD is also described (Evendon etal., 1993). Bacterial kidney disease lesions, involving extensivetissue infiltration by mononuclear immunocytes, granulomatouslesions, and necrotic foci, are typical of pathology associatedwith Type IV hypersensitivity (Flano et al., 1996).

Given the importance of vertical transmission in BKDepidemiology, BKD in juvenile salmonids and the ontogenetic changesassociated with disease pathogenesis are important considerations.This subject, however, has received comparatively little focus inthe literature. Immunological tolerance has been addressed as asignificant feature related to the ontogeny of the salmonid immunesystem and BKD; Brown et al.(1996) showed exposure of salmonideggs to p57 to be subsequently associated with diminished humoralimmune responses against p57 in fish hatched from these eggs. Fishinfected vertically and exposed to R. salmoninarum antigen prior tothe maturation of the immune system would likely be predisposed todevelop an immunological tolerance to the pathogen, impeding anyimmunological defense to be subsequently mounted.

-19-

G. References:Aedo, M.E. and S.P. Bustos. 1991. Production of salmon’s

eggs free of BKD. Chile-Pesq. 62:51-55.

Amend, D.F. and K.A. Johnson. 1984. Evidence for lack ofantigenic competition among various combinations of Vibrioangillarum, Yersinia ruckeri, Aeromonas salmonicida, andRenibacterium salmoninarum bacterins when administered tosalmonid fishes. Journal of Fish Diseases 7:293-299.

Armstrong, R.D., T.P.T. Evelyn, S.W. Martin, W. Dorward, andH.W. Ferguson. 1989. Erythromycin levels within eggs andalevins derived from spawning broodstock chinook salmonOncorhynchus tshawytscha injected with the drug. Diseases ofAquatic Organisms. 6:33-36.

Austin, B. 1985. Evaluation of antimicrobial compoundsfor the control of bacterial kidney disease in rainbow trout,Salmo gairdneri Richardson. Journal of Fish Diseases. 8:209-220.

Austin, B. 1983. Effectiveness of ozone for thedisinfection of laboratory effluent. FEMS Microbiology Letters. 19:21-214.

Austin, B. and J.N. Rayment. 1985. Epizootiology ofRenibacterium salmoninarum , the causative agent of bacterialkidney disease in salmonid fish. Journal of Fish Diseases. 8:505-509.

Balfry, S.K. 1990. Observations on the epizootiology ofbacterial kidney disease in farmed salmonids. Simon FrasierUniversity (Canada). Volume 31/01 of Masters Abstracts. 95 pp.

Bandin, I., C. Rivas, Y. Santos, C.J. Secombes, J.L. Barja,and A.E. Ellis. 1995. Effects of serum factors on thesurvival of Renibacterium salmoninarum within rainbow troutmacrophages. Diseases of Aquatic Organisms. 23(3):221-227.

Bandin, I., Y. Santos, C. Rivas, A.M. Skarmeta, and J.L.Barja. 1993a. Studies on the survival of Renibacteriumsalmoninarum within salmonid macrophages. 4th National Congresson Aquaculture, Vilanova de Arousa, Galacia, Spain. pp 21-24.

-20-

Bandin, I., A.E. Ellis, J.L. Barja, and C.J. Secombes.1993b. Interaction between rainbow trout macrophages andRenibacterium salmoninarum in vitro. Fish and ShellfishImmunology. 3:25-33.

Bandin, I., Y. Santos, J.L. Barja, and A.E. Toranzo. 1993c. Detection of a common antigen among Renibacteriumsalmoninarum , Corynebacterium aquaticum , and Carnobacteriumpiscicola by the Western blot technique. Journal of AquaticAnimal Health. 5(3):172-176.

Bandin, Y., Y. Santos, B. Magarinos, J.L. Barja, and A.E.Toranzo. 1992. The detection of two antigenic groups amongRenibacterium salmoninarum isolates. FEMS Microbiology Letter. 94:105-110.

Banner, C.R., J.J. Long, J.L. Fryer, and J.S. Rohovec. 1986. Occurrence of salmonid fish infected with Renibacteriumsalmoninarum in the Pacific ocean. Journal of Fish Diseases. 9:273-275.

Banner, C.R., J.S. Rohovec, and J.L. Fryer. 1983. Renibacterium salmoninarum as a cause of mortality among chinooksalmon in salt water. Journal of the World Mariculture Society. 14:236-239.

Beacham, T.D. and T.P.T. Evelyn. 1992. Population andgenetic variation in resistence of chinook salmon to vibriosis,furunculosis and bacterial kidney disease. Journal of AquaticAnimal Health. 4(3):153-167.

Belding, D.L. and B. Merrill. 1935. A preliminary reportupon a hatchery disease of the Salmonidae . Tranactions of theAmerican Fisheries Society. 65:76-84.

Bell, G.R., R.W. Hoffman, and L.L. Brown. 1990. Pathology of experimental infections of the sablefish, Anoplopomafimbria (Pallas), with Renibacterium salmoninarum , the agent ofbacterial kidney disease in salmonids. Journal of FishDiseases. 13:355-367.

-21-

Bell, G.R., G.S. Traxler, and C. Dworschak. 1988. Development in vitro and pathogenicity of an erythromycin-resistant strain of Renibacterium salmoninarum , the causativeagent of bacterial kidney disease in salmonids. Diseases ofAquatic Organisms. 4:19-25.

Bell, G.R., D.A. Higgs, and G.S. Traxler. 1984. Theeffect of dietary ascorbate, zinc, and manganese on thedevelopment of experimentally induced bacterial kidney disease insockeye salmon ( Oncorhynchus nerka ). Aquaculture. 36:293-311.

Bell, G.R. and G.S. Traxler. 1986. Resistance of thePacific lamprey, Lampropetra tridentata (Gairdner), to challengeby Renibacterium salmoninarum , the causative agent of bacterialkidney disease in salmonids. Journal of Fish Disease. 9(3):277-279.

Bowser, P.R., R.B. Landy, G.A. Wooster, and J.G. Babish. 1988. Efficacy of elevated dietary fluoride for the control ofRenibacterium salmoninarum infection in rainbow trout Salmogairdneri . Journal of the World Aquaculture Society. 19(1):1-7.

Brown, L.L., G.K. Iwama, and T.P.T. Evelyn. 1996. Theeffect of early exposure of coho salmon eggs to the p57 proteinof Renibacterium salmoninarum on the development of immunity tothe pathogen. Fish and Shellfish Immunology. 6:149-165.

Brown, L.L., G.K. Iwama, T.P.T. Evelyn, W.S. Nelson, andR.P. Levine. 1994. Use of the polymerase chain reaction (PCR)to detect DNA from Renibacterium salmoninarum within individualsalmonid eggs. Disease of Aquatic Organisms. 18:165-171.

Brown, L.L., L.J. Albright, and T.P.T. Evelyn. 1990. Control of vertical transmission of Renibacterium salmoninarum byinjection of antibiotics into maturing female coho salmonOncorhynchus kisutch . Diseases of Aquatic Organisms. 9:127-131.

Bruno, D.W. 1990. Presence of a saline extractableprotein associated with virulent strains of the fish pathogen,Renibacterium salmoninarum . Bulletin of the EuropeanAssociation of Fish Pathologists. 10(1):8-10.

Bruno, D.W. 1988. Bacterial kidney disease an increasingthreat to salmon. Fish Farmer. 11(6):15-17.

-22-

Bruno, D.W. 1987. Serum agglutinating titres againstRenibacterium salmoninarum the causative agent of bacterialkidney disease in rainbow trout, Salmo gairdneri Richardson, andAtlantic salmon, Salmo salar L. Journal of Fish Biology. 30:327-334.

Bruno, D.W. 1986a. Histopathology of bacterial kidneydisease in laboratory infected rainbow trout, Salmo gairdneriRichardson, and Atlantic salmon, Salmo salar L.,with reference tonaturally infected fish. Journal of Fish Diseases. 9:523-537.

Bruno, D.W. 1986b. Changes in serum parameters ofrainbow trout, Salmo gairdneri Richardson, and Atlantic salmon,Salmo salar L. infected with Renibacterium salmoninarum . Journal of Fish Diseases. 9:205-211.

Bruno, D.W. and R. Johnstone. 1990. Susceptibility ofdiploid and triploid Atlantic salmon, Salmo salar L., tochallenge by Renibacterium salmoninarum . Bulletin of theEuropean Association of Fish Pathologists. 10(2):45-47.

Bruno, D.W. and A.L.S. Munro. 1986. Haematologicalassessment of rainbow trout, Salmo gairdneri Richardson, andAtlantic salmon, Salmo salar L., infected with Renibacteriumsalmoninarum . Journal of Fish Diseases. 9:195-204.

Bullock, G.L. and R.L. Herman. 1988. Bacterial kidneydisease of salmonid fishes caused by Renibacterium salmoninarum . USDI Fish and Wildlife Service, Fish Disease Leaflet 78. 10 pp.

Bullock, G.L., B.R. Griffin, and H.M. Stuckey. 1980. Detection of Coryneform salmoninus by direct fluorescent antibodytest. Canadian Journal of Fisheries and Aquatic Science. 37:719-721.

Bullock, G.L. and H.M. Stuckey. 1975. Fluorescentantibody identification and detection of the Corynebacteriumcausing kidney disease of salmonids. Journal of the FisheriesResearch Board of Canada. 32:2224-2227.

Butendieck, B., P. Morales, J. Figueroa, M. Concha, and G.Leon. 1995. Detection of Renibacterium salmoninarum byspecific gene amplification. Archivos de Medicina Veterinaria. 27(2):47-54.

-23-

CaboBravo, C. 1993. Analysis of commercially farmedAtlantic salmon ( Salmo salar L. )by serological methods for thedetection of Renibacterium salmoninarum , the causative agent ofbacterial kidney disease. University of New Brunswick(Canada), Volume 33/02 of Masters abstracts. 93 pp.

Campbell, R., M. Thirty, and R.H. Richards. 1994. Passive immunization against Renibacterium salmoninarum. In: V.Vallyathan, V. Castranova, K. Weber, and N. Dalal (editors),Oxygen Radicals and Lung Injury, 1994 Volume, 102 pp.

Carlos, S.J. and J.L. Hackett. 1994. Relisa-BKD, a newrapid enzyme-linked immunosorbent assay for the detection ofRenibacterium salmoninarum antigen. International Symposium onAquatic Animal Health: Program and Abstracts. University ofCalifornia-Davis School of Veterinary Medicine. P. P-38.

Chein, M.S., T.L. Gilbert, C. Huang, M.L. Landolt, P.J.O’Hara, and J.R. Winton. 1992. Molecular cloning and sequenceanalysis of the gene coding for the 57-kDa major soluble antigenof the salmonid fish pathogen, Renibacterium salmoninarum . FEMSMicrobiology Letters. 96(2-3):259-266.

Chen, P.K., G.L. Bullock, H.M. Stuckey, and A.C. Bullock. 1974. Serological diagnosis of corynebacterial kidney diseaseof salmonids. Journal of the Fisheries Research Board ofCanada. 31:1939-1940.

Cipriano, R.C., C.E. Starliper, and J.H. Schachte. 1985. Comparative sensitivities of diagnostic procedures used to detectbacterial kidney disease in salmonid fishes. Journal ofWildlife Diseases. 21(2):144-148.

Cvitanich, J.D. 1994. Improvements in the directflourescent antibody technique for the detection, identification,and quantification of Renibacterium salmoninarum in salmonidsmears. Journal of Aquatic Animal Health. 6(1):1-12.

Daly, J.G. and R.M.W. Stevenson. 1989. Agglutination ofsalmonid spermatozoa by Renibacterium salmoninarum . Journal ofAquatic Animal Health. 1:163-164.

Daly, J.G. and R.M.W. Stevenson. 1987. Hydrophobicityand hemagglutinating properties of Renibacterium salmoninarum . Journal of General Microbiology. 133:3575-3580.

-24-

Dubreuil, D., M. Jacques, L. Graham, and R. Lallier. 1990. Identification and biochemical and structural characterizationof a fimbral hemagglutinin of Renibacterium salmoninarum . Journal of General Microbiology. 136:2443-2448.

Earp, B.J., C.H. Ellis, and E.J. Ordal. 1953. Kidneydisease in young salmon. State of Washington Department of Fisheries, Special Report Series No. 1. 73 pp.

Elliott, D.G. and T.Y. Barila. 1987. Membrane filtration flourescent antibody staining procedure for detecting andquantifying Renibacterium salmoninarum in coelomic fluid ofchinook salmon ( Oncorhynchus tshawytscha ). Canadian Journal ofFisheries and Aquatic Science. 44:206-209.

Elliott, D.G., R.J. Pascho, and A.N. Palmisano. 1995. Brood stock segregation for the control of bacterial kidneydisease can affect mortality of progeny chinook salmon( Oncorhynchus tshawytscha ) in seawater. Aquaculture. 132:133-144.

Elliott, D.G., R.J. Pascho, P.R. Manning, and C.L. McKibben. 1994. Studies on the potential for transmission ofRenibacterium salmoninarum to chinook salmon ( Oncorhynchustshawytscha ) during code-wire tagging procedures. InternationalSymposium on Aquatic Animal Health:Program and Abstracts. University of California - Davis School of Veterinary Medicine. pp. W-17.5

Elliott, D.G., R.J. Pascho, and G.L. Bullock. 1989. Developments in the control of bacterial kidney disease ofsalmonid fishes. Diseases of Aquatic Organisms. 6:201-215.

Evelyn, T.P.T., L. Prosperi-Porta, and J.E. Ketcheson. 1990. Two new techniques for obtaining consistent results whengrowing Renibacterium salmoninarum on KDM2 culture medium. Diseases of Aquatic Organisms. 9:209-212.

Evelyn, T.P.T., J.E. Ketcheson, and L. Prosperi-Porta. 1988. Trials with anti-bacterial kidney disease vaccines in twospecies of Pacific salmon. International Fish Health Conference(Vancouver, B.C.) Handbook, Fish Health Section, AmericanFisheries Society. p.38.

-25-

Evelyn, T.P.T., L. Prosperi-Porta, and J.E. Ketcheson. 1986a. Experimental intra-ovum infection of salmonid eggs withRenibacterium salmoninarum and vertical transmission of thepathogen with such eggs despite their treatment witherythromycin. Diseases of Aquatic Organisms. 1:197-202.

Evelyn, T.P.T., J.E. Ketcheson, and L. Prosperi-Porta. 1986b. Use of erythromycin as a means of preventing verticaltransmission of Renibacterium salmoninarum . Diseases of AquaticOrganisms. 2:7-11.

Evelyn, T.P.T., J.E. Ketcheson, and L. Prosperi-Porta. 1984. Further evidence for the presence of Renibacteriumsalmoninarum in salmonid eggs and for the failure of povidone-iodine to reduce the intra-ovum infection rate in water-hardenedeggs. Journal of Fish Diseases. 7:173-182.

Evelyn, T.P.T. 1977. An improved growth medium for thekidney disease bacterium and some notes on using the medium. Bulletin de L’Office international de Epizooties. 87(5-6):511-513.

Evendon, A.J., T.H. Grayson, M.L. Gilpin, and C.B. Munn. 1993. Renibacterium salmoninarum and bacterial kidney disease -the unfinished jigsaw. Annual Review of Fish Diseases, 1993. pp.87-104.

Evenson, O., O. Bendik, and A. Nilsen. 1994. Immunohistochemical identification of Renibacterium salmoninarumby monoclonal antibodies in paraffin-embedded tissues of Atlanticsalmon ( Salmo salar L.) Using paired immunoenzyme and pairedimmunofluorescence techniques. Journal of Veterinary DiagnosticInvestigation. 6:48-55.

Fiedler, F. and R. Draxl. 1986. Biochemical andimmunochemical properties of the cell surface of Renibacteriumsalmoninarum . Journal of Bacteriology. pp.799-804.

Flano, E., P. Lopez-Fierro, B. Razquin, S.L. Kaattari, andA. Villena. 1996. Histopathology of the renal and splenichaematopoietic tissues of coho salmon Oncorhynchus kisutchexperimentally infected with Renibacterium salmoninarum . Diseases of Aquatic Organisms. 24:1076-115.

-26-

Fryer, J.L. and J.E. Sanders. 1981. Bacterial kidneydisease of salmonid fish. Annual Review of Microbiology. 35:273-298.

Getchell, R.G., J.S. Rohovec, and J.L. Fryer. 1985. Comparison of Renibacterium salmoninarum isolates by antigenicanalysis. Fish Pathology. 20(2/3):149-159.

Gjedrem, T. and H.M. Gjoen. 1995. Genetic variation insusceptibility of Atlantic salmon, Salmo salar L., tofurunculosis, BKD, and cold water vibriosis. AquacultureResearch. 26(2):129-134.

Griffiths, S.G. and W.H. Lynch. 1991. Instability of themajor soluble antigen produced by Renibacterium salmoninarum . Journal of Fish Diseases. 14:55-66.

Griffiths, S.G., G. Olivier, J. Fildes, and W.H. Lynch. 1991. Comparison of western blot, direct fluorescent antibodyand drop-plate culture methods for the detection of Renibacteriumsalmoninarum in Atlantic salmon ( Salmo salar L.). Aquaculture. 97:117-129.

Hardie, L.J., A.E. Ellis, and C.J. Secombes. 1996. Invitro activation of rainbow trout macrophages stimulatesinhibition of Renibacterium salmoninarum growth concomitant withaugmented generation of respiratory burst products. Diseases ofAquatic Organisms. 25:175-183.

Hariharan, H., B. Qian, B. Despres, F.S. Kibenge, S.B.Heaney, and D.J. Rainnie. 1995. Development of a specificbiotinylated DNA probe for the detection of Renibacteriumsalmoninarum . Canadian Journal of Veterinary Research. 58(4):306-310.

Hayakawa, Y., T. Harada, K. Hatai, S.S. Kubota, T. Bunya,and G. Hoshiai. 1989. Histopathology of BKD (Bacterial kidneydisease) occurred in Sea-cultured Coho salmon ( Oncorhynchuskisutch ). Fish Pathology. 24(1):17-21.

Hicks, B.D. and J.R. Geraci. 1984. A histologicalassessment of damage in rainbow trout, Salmo gairdneriRichardson, fed rations containing erythromycin. Journal ofFish Diseases. 7:457-465.

-27-

Hsu, H.M., G.A. Wooster, and P.R. Bowser. 1994. Efficacyof enrofloxacin for the treatment of salmonids with bacterialkidney disease, caused by Renibacterium salmoninarum . Journalof Aquatic Animal Health. 6(3):220-223.

Hsu, H.M., P.R. Bowser, and J.H. Schachte Jr. 1991. Development and evaluation of a monoclonal-antibody-based enzyme-linked immunosorbent assay for the diagnosis of Renibacteriumsalmoninarum infection. Journal of Aquatic Animal Health. 3:168-175.

Iwama, G.K. and G.L. Greer. 1980. Effect of a bacterialinfection on the toxicity of sodium pentachlorophenate tojuvenile coho salmon. Transactions of the American FisheriesSociety. 109:290-292.

Jansson, E., T. Hongslo, R. Lindberg, O. Ljungberg, and B.M.Svensson. 1991. Detection of Renibacterium salmoninarum andYersinia ruckeri by the peroxidase-antiperoxidaseimmunohistochemical technique in melanin-containing cells of fishtissues. Journal of Fish Diseases. 14:689-692.

Kaattari, S.L., P. Turaga, and G. Wiens. 1989. Development of a vaccine for bacterial kidney disease in salmon -final report. U.S. Department of Energy, Bonneville PowerAdministration, Division of Fish and Wildlife, Portland, OR. 310pp.

Kettler, S., C. Pfeil-Putzien, and R. Hoffmann. 1986. Infection of grayling ( Thymallus thymallus ) with the agent ofbacterial kidney disease (BKD). Bulletin of the EuropeanAssociation of Fish Pathologists. 6(3):69-71.

Kimura, T, Y. Ezura, K. Tajima, and M. Yoshimizu. 1978. Serological diagnosis of bacterial kidney disease of salmonid(BKD): immunodiffusion test by heat stable antigen extracted frominfected kidney. Fish Pathology. 13(2):103-108.

Klein, T.C. 1992. Intracellular survival andmultiplication of Renibacterium salmoninarum within culturedmacrophages from rainbow trout ( Oncorhynchus mykiss ). University of Prince Edward Island, Volume 32/01 of MastersAbstracts. 181 pp.

-28-

Lall, S.P., W.D. Paterson, J.A. Hines, and N.J. Adams. 1985. Control of bacterial kidney disease in Atlantic salmon,Salmo salar L., by dietary modification. Journal of FishDiseases. 8:113-124.

Lee, E.G.H. and T.P.T. Evelyn. 1994. Prevention ofvertical transmission of the bacterial kidney disease agentRenibacterium salmoninarum by broodstock injection witherythromycin. Diseases of Aquatic Organisms. 18(1):1-4.

Leon, G., M.A. Martinez, J.P. Etchegaray, M.I. Vera, J.Figueroa, and M. Krauskopf. 1994a. Specific DNA probes forthe identification of the fish pathogen, Renibacteriumsalmoninarum . World Journal of Microbiology and Biotechnology. 10:149-153.

Leon, G., N. Maulen, J. Figueroa, J. Villanueva, C.Rodriguez, M. I. Vera, and M. Krauskopf. 1994b. A PCR-basedassay for the identification of the fish pathogen Renibacteriumsalmoninarum . FEMS Microbiology Letters. 115:131-136.

Lovely, J.E., C. Cabo, S.G. Griffiths, and W.H. Lynch. 1994. Detection of Renibacterium salmoninarum infection inasymptomatic Atlantic salmon. Journal of Aquatic Animal Health. 6(2):126-132.

Magnusson, H.B., O.H. Fridjonsson, O.S. Andresson, E.Benediktsdottir, S. Gudmundsdottir, and V.Andresdottir. 1994. Renibacterium salmoninarum , the causative agent of bacterialkidney disease in salmonid fish, detected by nested reversetranscription-PCR of 16S rRNA sequences. Applied andEnvironmental Microbiology. 60(12):4580-4583.

McCarthy, D.H., T.R. Croy, and D.F. Amend. 1984. Immunization of rainbow trout, Salmo gairdneri , Richardson,against bacterial kidney disease:preliminary efficacy evaluation. Journal of Fish Diseases. 7:65-71.

Mitchum, D.L. and L.E. Sherman. 1981. Transmission ofbacterial kidney disease from wild to stocked hatchery trout. Canadian Journal of Fisheries and Aquatic Science. 38:547-551.

-29-

Mitchum, D.L., L.E. Sherman, and G.T. Baxter. 1979. Bacterial kidney disease in feral populations of brook trout( Salvelinus fontinalis ) and rainbow trout ( Salmo gairdneri ). Journal of the Fisheries Research Board of Canada. 36:1370-1376.

Moffitt, C.M. 1992. Survival of juvenile chinook salmonchallenged with Renibacterium salmoninarum and administered oraldoses of erythromycin thiocyanate for different durations. Journal of Aquatic Animal Health. 4(2):119-125.

Moffitt, C.M. 1991. FDA approved registration oferythromycin for treatment of bacterial kidney disease (BKD) injuvenile and adult chinook salmon. U.S. Department of Energy -Annual Report, 1989-1990. 54 pp.

Olivier, G., S.G. Griffiths, J. Fildes, and W.H. Lynch. 1992. The use of western blot and electroimmunotransfer blotassays to monitor bacterial kidney disease in experimentallychallenged Atlantic salmon, Salmo salar L. Journal of FishDiseases. 15:229-241.

Ordal, E.J. and B.J. Earp. 1956. Cultivation andtransmission of etiological agent of kidney disease in salmonidfishes (22392). Proceedings of the Society for ExperimentalBiology and Medicine. 92:85-88.

Ortega, C., F. Sanz, J.L. Muzquiz, P. Ramos, J. Docando, andE. Planas. 1994. Epidemiological study of BKD ( Renibacteriumsalmoninarum ) in Spain: role of temperature as a risk factor in anatural outbreak of the disease. Bulletin of the EuropeanAssociation of Fish Pathologists. 13(6)201-202.

Paclibare, J.O., T.P.T. Evelyn, L.J. Albright, and L.Prosperi-Porta. 1994. Clearing of the kidney diseasebacterium Renibacterium salmoninarum from seawater by the bluemussel Mytilus edulis , and the status of the mussel as areservoir of the bacterium. Diseases of Aquatic Organisms. 18(2):129-133.

Pascho, R.J., D.G. Elliot, and J.M. Streufert. 1991. Brood stock segregation of spring chinook salmon Oncorhynchustshawytscha by use of the enzyme-linked immunosorbent assay(ELISA) and the fluorescent antibody technique (FAT) affects theprevalence and levels of Renibacterium salmoninarum infection inprogeny. Diseases of Aquatic Organisms. 12:25-40.

-30-

Pascho, R.J., M.L. Landolt, and J.E. Ongerth. 1995. Inactivation of Renibacterium salmoninarum by free chlorine. Aquaculture. 131(3-4):165-175.

Pascho, R.J. and D. Mulcahy. 1987. Enzyme-linkedimmunosorbent assay for a soluble antigen of Renibacteriumsalmoninarum , the causative agent of salmonid bacterial kidneydisease. Canadian Journal of Fisheries and Aquatic Science. 44:183-191.

Paterson, W.D., D. Desautels, and J.M. Weber. 1981a. Theimmune response of Atlantic salmon, Salmo salar L., to thecausative agent of bacterial kidney disease, Renibacteriumsalmoninarum . Journal of Fish Diseases. 4:99-111.

Paterson, W.D., S.P. Lall, and D. Desautels. 1981b. Studies on bacterial kidney disease in Atlantic salmon ( Salmosalar ) in Canada. Fish Pathology. 15(3/4):283-292.

Piganelli, J.D. 1995. Development of enteric protectedvaccines for aquaculture ( Renibacterium salmoninarum ). OregonState University, Volume 56/10-B of Dissertation AbstractsInternational. 159 pp.

Radacovici, S. and J.D. Dubreuil. 1991. Amino-terminalsequencing and antigenic analysis of a 57 kD protein fromRenibacterium salmoninarum . Biomedical Letters. 46:219-225.

Romalde, J.L., B. Magarinos, B. Fouz, I. Bandin, S. Nunez,and A.E. Toranzo. 1995. Evaluation of BIONOR Mono-kits forrapid detection of bacterial fish pathogens. Diseases ofAquatic Organisms. 21(1):25-34.

Rose, A.S. and R.P. Levine. 1992. Complement-mediatedopsonization and phagocytosis of Renibacterium salmoninarum . Fish and Shellfish Immunology. 2(3):223-240.

Sakai, M., N. Amaaki, S. Atsuta, and M. Kobayashi. 1987. Comparative sensitivity of dot blot methods used to detectRenibacterium salmoninarum . Journal of Fish Diseases. 10:415-418.

-31-

Sakai, M., S. Atsuta, and M. Kobayashi. 1993. The immuneresponse of rainbow trout ( Oncorhynchus mykiss ) injected withfive Renibacterium salmoninarum bacterins. Aquaculture. 113:11-18.

Sakai, M., S. Atsuta, and M. Kobayashi. 1991. Susceptibility of five salmonid fishes to Renibacteriumsalmoninarum . Fish Pathology. 26(3):159-160.

Sakai, M., S. Atsuta, and M. Kobayashi. 1989. Attemptedvaccination of rainbow trout Oncorhynchus mykiss againstbacterial kidney disease. Nippon Suisan Gakkaishi. 55(12):2105-2109.

Sakai, M. K. Miyama, S. Atsuta, and H. Kobayashi. 1996. The chemiluminescent responses of phagocytic cells of cohosalmon, rainbow trout, and carp against Renibacteriumsalmoninarum . Fish and Shellfish Immunology. 6:71-73.

Sami, S., T. Fischer-Scherl, R.W. Hoffman, and C. Pfeil-Putzien. 1992. Immune complex-mediated glomerulonephritisassociated with bacterial kidney disease in the rainbow trout( Oncorhynchus mykiss ). Veterinary Pathology. 29:169-174.

Sanders, J.E. and J.L. Fryer. 1980. Renibacteriumsalmoninarum gen. nov., the causative agent of bacterial kidneydisease in salmonid fishes. International Journal of SystematicBacteriology. 30(2):496-502.

Sanders, J.E., K.S. Pilcher, and J.L. Fryer. 1978. Relation of water temperature to bacterial kidney disease in cohosalmon ( Oncorhynchus kisutch ), sockeye salmon ( O. nerka ), andsteelhead trout ( Salmo gairdneri ). Journal of the FisheriesResearch Board of Canada. 35:8-11.

Senson, P. and R.M.W. Stevenson. 1994. Surfacecomparisons of two strains of the fish pathogen Renibacteriumsalmoninarum . International Symposium on Aquatic AnimalHealth:Programs and Abstracts. University of California-DavisSchool of Veterinary Medicine. p.W-14.3.

Smith, I.W. 1964. The occurrence and pathology of Deedisease. Freshwater Salmon Fisheries Research. 34:3-12.

-32-

Speare, D.J., V.E. Ostland, and H.W. Ferguson. 1993. Pathology associated with meningoencephalitis during bacterialkidney disease of salmonids. Research in Veterinary Science. 54:25-31.

Stackebrandt, E., U. Wehmeyer, H. Nader, and F. Fiedler. 1988. Phylogenetic relationship of the fish pathogenicRenibacterium salmoninarum to Arthrobacter, Micrococcus andrelated taxa. FEMS Microbiology Letters. 50:117-120.

Suzumoto, B.K., C.B. Schreck, and J.D. McIntyre. 1977. Relative resistances of three transferrin genotypes of cohosalmon ( Oncorhynchus kisutch ) and their hematological responsesto bacterial kidney disease. Journal of the Fisheries ResearchBoard of Canada. 34(1):1-8.

Teska, J.D. 1994. In vitro growth of the bacterialkidney disease organism Renibacterium salmoninarum on a nonserum,noncharcoal-based “homospecies-metabolite” medium. Journal ofWildlife Diseases. 30(3):383-388.

Teska, J.D., A. Dawson, C.E. Starliper, and D. Tillinghast. 1995. A multiple-technique approach to investigating thepresumptive low-level detection of Renibacterium salmoninarum ata broodstock hatchery in Maine. Journal of Aquatic AnimalHealth. 7(3):251-256.

Thorarinsson, R., M.L. Landolt, D.G. Elliot, R.J. Pascho,and R.W. Hardy. 1994. Effect of dietary vitamin E andselenium on growth, survival, and the prevalence of Renibacteriumsalmoninarum infection in chinook salmon ( Oncorhynchustshawytscha ). Aquaculture. 121(4):343-358.

Turaga, P., G. Wiens, and S. Kaattari. 1987. Bacterialkidney disease:the potential role of soluble protein antigen(s). Journal of Fish Biology. 31 (Supplement A):191-194.

Wiens, G.D. and S.L. Kaattari. 1991. Monoclonal antibodycharacterization of a leukoagglutinin produced by Renibacteriumsalmoninarum . Infection and Immunity. 59(2):631-637.

Wiens, G.D. and S.L. Kaattari. 1989. Monoclonal antibodyanalysis of common surface protein(s) of Renibacteriumsalmoninarum . Fish Pathology. 24(1):1-7.

-33-

White, M.R., W. Ching-Ching, and S.R. Albregts. 1995. Comparison of diagnostic tests for bacterial kidney disease injuvenile steelhead trout ( Oncorhynchus mykiss ). Journal ofVeterinary Diagnostic Investigation. 7(4):494-499.

Winter, G.W., C.B. Schreck, and J.D. McIntyre. 1980. Resistance of different stocks and transferrin genotypes of cohosalmon, Oncorhynchus kisutch , and steelhead trout, Salmogairdneri , to bacterial kidney disease and vibriosis. FisheryBulletin. 77(4):795-802.

Wolf, K. and C.E. Dunbar. 1959. Methods of infectingtrout with kidney disease and some effects of temperature onexperimental infections. U.S.D.I., Fish and Wildlife Service,Special Scientific Report-Fisheries No. 286. 8 pp.

Wood, P.A. and S.L. Kaattari. 1996. Enhancedimmunogenicity of Renibacterium salmoninarum in chinook salmonafter removal of the bacterial cell surface-associated 57 kDaprotein. Diseases of Aquatic Organisms. 25:71-79.

Wood, J.W. and J. Wallis. 1955. Kidney disease in adultchinook salmon and its transmission by feeding to young chinooksalmon. Research briefs, Fish Commission of Oregon, Portland,OR. 6(2)32-40.

Young, C.L. and G.B. Chapman. 1978. Ultrastructuralaspects of the causative agent and renal histopathology ofbacterial kidney disease in brook trout ( Salvelinus fontinalis ). Journal of the Fisheries Research Board of Canada. 35:1234-1248.

-34-

` Afterword/Introduction to the Chapters:

With current limitations on the use of chemoprophylactics andtherapeutics in aquatic species, it is imperative that we look toother means to achieve successful control of BKD. Management/husbandry techniques and vaccination rank foremost among thealternatives. To accomplish this, we first need to expand ourknowledge base concerning the immunology of BKD in order that wemay better manipulate these immune responses to our advantage.There are a number of aspects of BKD immunology which warrantfurther investigation, and a number of these specific areas areaddressed in the following chapters of this dissertation.

In general, these chapters address the issues of BKD and thesalmonid immune response in terms of:

1- The elucidation of the relationship of specific environmentalfactors to BKD pathogenesis.

Environmental parameters may physiologically depressimmunological function and thereby increase a fish’s susceptibilityto diseases. As previously mentioned, many exogenous factors whichinfluence BKD development in feral and/or cultured have beenidentified and may be controlled in salmonid husbandry to reducethe severity of BKD. High levels of morbidity and mortality arereported to occur among cultured juvenile salmonids infectedvertically with BKD, and the extent to which management techniquemay reduce BKD-related losses in these stocks warrants furtherinvestigation. Chapter 1 of this dissertation examines the effectsof two environmental conditions common to fish culture,overcrowding and overfeeding, as they effect BKD development and R.salmoninarum antigen prevalence among juvenile chinook salmon.

2- The identification of immunomodulatory factors involved in BKDpathogenesis.

A more precise determination of the immunomodulatorymechanisms involved in the interaction between R. salmoninarum andits salmonid host is needed. In particular, the immunologicalparameters which are affected and the role of the R. salmoninarumECP in these interactions are of interest. Chapter 2 of thisdissertation examines the in vivo immune response of juvenilerainbow trout following exposure to the ECP in terms of bothhumoral and cell-mediated immunological parameters, including theability to mount a secondary immune response against anotherbacterial pathogen. Chapter 3 addresses the in vitro effects ofthe ECP upon splenic immunocyte functions, phagocytosis andrespiratory burst activity, in brook trout.

-35-

3- The definition of hypersensitivity reactions in BKD.Immune-mediated hypersensitivity, particularly immune-complex-

mediated hypersensivity, in BKD has considerable ramifications forcontrol measures involving immunostimulation via antigen exposure.Further investigation is warranted in order to discern theimportance and consistency of immunological hypersensivity in BKDpathogenesis. Chapter 4 of this dissertation examines renallesions, including immunopathology and indications of immune-mediated disease, of brook trout exposed to R. salmoninarum throughthe consumption of infected tissues.

Most assuredly, there is an extensive list of BKD/immunologyrelated topics which require further elucidation. It is my hopethat these chapters will contribute to the overall research effortswhich are attempting to better discern the relationship betweenbacterial kidney disease and the salmonid immune response; thus, wemay better understand, and ultimately gain a greater measure ofcontrol over, this disease among cultured and feral salmonids.

-36-

Chapter 1

Effects of stocking density and feeding rate uponRenibacterium salmoninarum prevalence and

bacterial kidney disease among chinook salmon( Oncorhynchus tshawytscha ).

-37-

Abstract

Two groups of chinook salmon( Oncorhynchus tshawytscha ) fingerlings,the progeny of brood stock infected either asymptomatically orclinically with Renibacterium salmoninarum , were maintained fortwelve weeks in nine experimental groups representing threedifferent levels of stocking density (24 g/l, 40 g/l, and 48 g/l)and feeding rate (2.8%, 5.6%, and 8.4% body weight/day). Prior toand during the investigation, renal samples from fish from eachgroup were evaluated for R. salmoninarum antigen prevalence byenzyme-linked immunosorbent assays (ELISA) and direct fluorescentantibody tests (DFAT). Water quality parameters, includingdissolved oxygen and unionized ammonia levels, were monitored inall groups, and mortality and growth rate data were collectedthroughout the twelve weeks. No significant differences in antigenprevalence were detected among the experimental groups for eachprogeny group, indicating a lack of effect of the potentialstressors related to stocking density and feeding rate in thisinvestigation. There was, however, a significant difference inantigen prevalence between the two progeny groups. For the progenyof asymptomatic fish, there was an initial antigen prevalenceproportion of 0.10 which decreased over the twelve weeks. For theprogeny of clinically diseased fish, the initial proportion ofprevalence was 0.50, increasing to 0.75 over the twelve weeks. Forthe latter progeny group, survival and growth rate analysesindicated greater survival and growth for fish exposed to thehigher density/feeding rate combinations. Significant correlationbetween density level and water quality parameters was evident inboth progeny groups. Low correlation was found for ELISA and DFATresults.

-38-

I. IntroductionRenibacterium salmoninarum , the etiological agent of bacterial

kidney disease (BKD) of salmonid fish, is a small (0.5 x 1.0 µm)Gram-positive diplobacillus which infects cultured and feralsalmonids. Pathogenesis is most often chronic and multisystemic,involving granulomatous lesions and widespread bacterialdissemination with intracellular localization. Disease transmissionoccurs horizontally from infected to uninfected individuals or viathe ingestion of infected tissues. Intra-ovum transmission of thebacterium appears to be the primary mode of infection in the wild(Evendon et al., 1993) and is similarly of considerable importancein cultured fish (Bruno and Munro, 1986; Evelyn et al., 1986).

Control of BKD has been attempted with varying degrees ofsuccess via chemical prophylaxis/therapeutics and throughmanagement strategies. Many aspects of salmonid culture managementcan influence R. salmoninarum prevalence and clinical developmentamong stocks. For instance, dietary components such as traceminerals and ascorbate have been shown to affect salmonidresistance to BKD (Elliott et al., 1989; Bowser et al., 1988; Lallet al., 1985). Suzimoto et al. (1977) also demonstrated thatgenetic manipulation of fish strains may directly affect theirsusceptibility to BKD. Research strongly supports brood stocksegregation based upon R. salmoninarum antigen prevalence as amanagement technique which may significantly reduce the prevalenceof R. salmoninarum in progeny and subsequently decrease the levelof BKD-related mortality (Evelyn et al., 1986; Elliott et al,1995). Given the limited scope of prevention and treatment optionscurrently available for control of BKD in cultured fish, it isimperative that husbandry related factors be identified and managedto decrease the impact of BKD among salmonid stocks.

Severity of BKD is affected by environmental variables such aswater temperature, salinity, and ionic composition (Wolf andDunbar, 1959; Sanders et al., 1978; Evendon et al., 1993). Theseeffects may be exerted through effects upon the pathogen, thehost's immune response, or a combination of the two. Two otherenvironmental variables that might influence the presence andseverity of BKD are stocking density and feeding rate. Overcrowding of fish is a common problem in hatcheries and may bedirectly detrimental as it influences social interaction andphysiological responses. High stocking densities/overcrowding havebeen associated with physiological alterations in salmonidsincluding decreased thyroid hormone levels, decreased plasmaprotein levels, altered plasma cortisol levels, and decreasedgrowth rates (Leatherland and Cho, 1985; Laidley and Leatherland,1988; Pickering and Stewart, 1984). The correlation between highstocking densities and increased mortality is another indicatorthat overcrowding is a significant environmental influence

-39-

(Fagerlund et al., 1981; Whalen and LeBar, 1994). Overcrowding maybe indirectly problematic as it contributes to poor water quality,namely through increased levels of nitrogenous waste and decreaseddissolved oxygen content.

Overfeeding is another potential environmental parameter whichmay influence BKD development. Similar to the effects ofovercrowding, overfeeding produces excessive amounts of organicdebris as sediment and particulate matter in the water. Waterquality parameters such as dissolved oxygen content, nitrogenouswaste levels, and total suspended solids may be secondarily alteredto the detriment of the fish.

In this study, the relationship between two environmentalvariables, stocking density and feeding rate, and BKD-relatedresponse variables including R. salmoninarum antigen prevalence,growth rate, and survival was examined in two populations of R.salmoninarum -infected chinook salmon ( Oncorhynchus tshawytscha )fingerlings.

II. Materials and MethodsFertilized eggs from two populations of chinook salmon from

Lake Ontario and Lake Michigan were shipped to the National FishHealth Research Laboratory (Kearneysville, WV). Brood stock thatproduced eggs from Lake Ontario were previously diagnosed positivefor R. salmoninarum through serological analysis, although noevidence of clinical BKD was present. These asymptomatic broodstock were tested for R. salmoninarum by the enzyme-linkedimmunosorbent assay (ELISA) with 45% of fish being positive and bythe direct fluorescent antibody test (DFAT) with 1.7% of fish beingpositive. The brood stock that spawned the eggs from Lake Michiganwere clinically affected with BKD as evidenced by lesions in kidneyand spleen tissues. Eggs were hatched in Heath incubators, and frywere reared in troughs (total volume= 415 l) supplied with springwater (12.5 C, total hardness= 240 ppm). Fish were maintained ono

a diet of commercial salmonid growth feed (Zeigler Brothers, Inc.,Gardners, PA). Seven months after hatching, 100 fingerlings fromthe subclinically infected brood stock (average weight=3.9g/fingerling) and fifty fingerlings from the clinically infectedbrood stock (average weight= 7.9 g/fingerling) were sacrificed.Kidney tissues were aseptically removed from all specimens andfrozen at -70 C. For the clinically affected progeny, all renalo

samples were additionally utilized to prepare tissue smears onmicroscope slides.

For each population (subclinically and clinically infectedprogeny groups), nine treatment groups with three replicates eachwere established (Table 1). Fingerlings were arbitrarily

-40-

distributed to each of 27 circular flow-through holding tanks (35l capacity) with a water flow rate of 1-2 l/minute. Nine tankswere stocked at each of the following density levels: 24 g fish/l(1.5 lbs/cubic foot); 40 g/l (2.5 lbs/cubic foot); 48 g/l (3.0lbs/cubic foot). Densities were based on the premise that 24 g/lrepresents near ideal rearing density (personal communication, AndyGreulich, manager of Salmon River State Fish Hatchery) for culturedchinook salmon fingerlings, and therefore, the two higher densitiesrepresented two levels of overcrowding. For each density, threegroups were fed at one of three rates: 2.8% body weight per day;5.6 % body weight per day; 8.4% body weight per day. The 2.8% wasconsidered optimal (Piper et al., 1982), and the additional levelsrepresented overfeeding. Fish were fed three times daily with apelleted commercial salmon diet (Zeigler Brothers, Inc., Gardners,PA). Water quality parameters including temperature, pH, dissolvedoxygen content, and total ammonia were monitored and recorded twoto three times per week. Temperature and dissolved oxygenmeasurements were recorded using a dissolved oxygen meter (YSI,Yellow Springs, OH), and pH was determined with a standard pH meter(Corning Model 220, Corning, NY). A colorimetric method(Wedemeyer and Yasutake, 1977) was employed to determine totalammonia concentration in water samples. Tanks were uniformlycleaned, generally every other day, by draining and refillingapproximately 2/3 of the water volume. At each feeding and waterquality sampling, fish were observed for general behavior andphysical lesions, mortalities were recorded, and dead fish werecollected from the experimental tanks.

Experimental tanks for both the subclinically and clinicallyaffected progeny groups were maintained for twelve weeks. At 4, 8,and 12 weeks, all fish from each experimental tank were weighedusing a digital scale (Yamamoto Corp. Model DPI-320, ColoradoSprings, CO) to determine growth rates. Ten to twenty-five fishwere removed from each tank for tissue sampling at the same fourweek intervals; the number of fish removed per tank was chosen toreadjust the density to the initial density level. Fish weresacrificed using a lethal dose of tricaine methanesulfonate (MS-222; Sigma Chemical Co., St. Louis, MO). Renal tissue samples wereaseptically removed and placed individually into labeled, pre-weighed 2.0 ml polypropylene vials (Costar Biofreeze; Cambridge,MA). Vials were reweighed and frozen at -70 C. Additionally,o

renal tissues from the clinical progeny groups were used to preparecytological specimens for direct fluorescent antibody testing(DFAT; Bullock et al., 1980).

Renibacterium salmoninarum antigen prevalence in all renaltissues collected from sampled fingerlings and mortalities fromboth progeny groups was assessed by ELISA. Tests were performedusing a modification of the technique described by Pascho and

-41-

Mulcahy (1987). Frozen renal tissue samples collected throughoutthe experiment were thawed at 4 C and diluted 1:15 (weight:volume)o

with sterile phosphate buffered saline containing 0.05% Tween 20(PBS-T20; pH=7.4). Tissue solutions were emulsified for 30 sec.with a sterile pipette, heated to 100 C for 15 min., ando

centrifuged at 8800 xg for 10 minutes. Samples were stored at 4 C. o

All ELISA buffers, antibody, antigen, reagents, and substrateswere obtained from Kirkegaard & Perry, Gaithersburg, MD. Flat-bottomed microtiter plates (96 wells; Nunc-Immuno Plate, Nunc,Neptune, NJ) were coated with a 1:10 solution of coating bufferwith 1:2000 coating antibody (affinity purified antibody to R.salmoninarum ). No coating antibody was used in substrate controlwells. Two hundred microliters of this solution was dispensed intoeach well of a microtiter plate, and the plates were covered withan adhesive seal and stored for 18 h. in a humidified chambers at4 C. Plates were then washed with a washing buffer, and 50 µl ofo

each sample was dispensed into the microplate wells in triplicate.Positive control antigen and negative controls prepared from renaltissue of R. salmoninarum -free trout were processed along with allsamples, and plates were re-sealed and incubated for 3 h. at 25 C. o

Following incubation, plates were washed and 200 µl of aperoxidase-conjugated antiserum (peroxidase-labeled affinitypurified antibody to R. salmoninarum diluted 1:200 in milkdiluent/blocking solution was added to all wells. Plates wereincubated for an additional 2 h. at 25 C. Plates were washed againo

and 200 µl of ABTS peroxidase substrate was added to each plate.Following an additional 20 min. of incubation, the reaction wasterminated with 50 µl of stop solution added to each well (ABTSperoxidase stop solution). Plates were evaluated for opticaldensity using a Microplate Bio-kinetics Reader (BioTek InstrumentsInc. EL 312e, Winooski, VT) and an IBM compatible computer.Optical density (OD) readings were recorded for each well, and theaverage OD reading of three wells was calculated for each sample.Negative control sample values were averaged and the standarddeviation from the mean calculated. A positive-negative thresholdfor samples was determined based upon the average negative controlOD value plus two standard deviations from the mean. Each samplewas subsequently assigned as either a "positive" or "negative"ELISA result based upon these calculations.

Additionally, renal tissues from experimental samples andmortalities from the clinical progeny group were examined by theDFAT. Briefly, the cytological slide preparations from renaltissues were coated with 1-2 drops of a solution containing 1:40FITC-labeled antiserum (fluorescein-labeled affinity purifiedantibody to R. salmoninarum ; Kirkegaard & Perry) and 1:100Rhodamine B in phosphate buffered saline (PBS). Slides wereincubated at 25 C for 20 min., rinsed in PBS, and rapidly air-o

-42-

dried. Dried slides were covered with a coverslip using mountingfluid and viewed under oil immersion with a fluorescent microscope.Fluorescence with the morphology of R. salmoninarum cells wasindicative of a positive response. Following the count of 100microscope fields, samples were categorized numerically:0=negative; 1= 1-5 cells per field; 2= >5 cells per field.

Statistical analysis of results was performed using SASprograms for Kaplain-Meier survival analysis for survival data,multiple linear regression for growth rates and water qualitycovariancy data (i.e.- dissolved oxygen and unionized ammonialevels), and logistic regression for ELISA and DFAT evaluation ofantigen prevalence (SAS PROCLIFE, SAS PROCGLM, SAS PROCLOGISTIC;SAS Institute, Inc., Cary, NC). Additionally, agreement betweenELISA and DFAT results for the antigen prevalence among treatmentlevels of the clinical progeny group was assessed using the Kappastatistic of correlation (Agresti, 1990).

III. Results

Antigen prevalence difference between progeny groupsThe overall proportion (i.e.- not considering treatment group

specificity) of fingerlings testing positive for R. salmoninarumantigen via ELISA was significantly different for the progeny ofthe subclinically infected brood stock versus the progeny of theclinically infected brood stock (referred to as "subclinicalprogeny" and "clinical progeny" for brevity) for all samplingintervals. The proportion of ELISA positive fingerlings sampledwas 0.1 for the subclinical progeny at time=0, decreasing to nearly0 for the 4, 8, and 12 week sample (4 weeks=0.034; 8 weeks=0.010;12 weeks=0.001). For the clinical progeny group, the initial(time=0) sampling proportion was 0.5, rising to a mean of 0.73 atthe 4 week sample, 0.72 at the 8 week sample, and 0.75 at the 12week sample (Figure 1).

A wide range in ELISA OD values was found among positivesamples from the clinical progeny group. Typically, negativethreshold values for ELISA ranged from 0.10 to 0.15 for subclinicaland clinical progeny samples. The OD values for clinical progenysamples testing positive ranged from slightly above the negativethreshold (approximately 0.20) to off-scale values (>3.00).Colorimetric change was generally visible to the naked eye at ODvalues of >1.00. Renal samples from mortalities (clinical progenygroup) all produced OD values at the higher end of the range(>1.00); most of these values were off-scale. The few positivesamples from the subclinical progeny group had OD values in thelower end of the range (<0.50). In addition, ELISA testing ofrenal tissues from the clinical progeny brood stock supported the

-43-

observation that clinical BKD was present in the population; fiveof the six samples tested strongly positive (i.e. intense colorchange apparent/OD>2.00) for R. salmoninarum antigen.

Mortality/survival among the progeny groupsDifferences in mortality rates between the two progeny groups

were appreciable. Sixteen mortalities were recorded cumulativelyfor all experimental groups of the subclinical progeny, and none ofthese fish tested positive for R. salmoninarum by ELISA. Incontrast, 1,218 mortalities were recorded for the experimentalgroups of the clinical progeny, and all of these tested positive byELISA and/or DFAT. Mortalities for the clinical progeny group aresummarized weekly in Table 2. Generally, mortality rates for allexperimental groups were highest during the first few weeks of theexperiment, dropping off considerably during the latter weeks.Diseased and dead fish showed similar clinical signs in theaffected progeny group. These signs included petecchial hemorrhage(externally and internally), exophthalmia, hyperpigmentation,hyphema, distension of the body cavity with free serosanguinousfluid accumulation, and focal or diffuse purulent exudate ofinternal organs (particularly liver, spleen, and kidney).

Data acquired from mortality attributed to BKD via positiveserological test results was used to formulate survival curves forthe clinical progeny groups (Figure 2). Fingerling survival wassimilar for the five treatment groups with lower combinations ofdensity and feeding levels (D1F1, D1F2, D1F3, D2F1, D2F2) such thatthe probability of fingerling survival after 12 weeks rangedbetween 0.55 and 0.57. For the remaining four treatment groups(D2F3, D3F1, D3F2, D3F3), the probability of fingerling survivalwas somewhat higher, ranging from 0.63 to 0.69.

Due to the absence of BKD-related mortality among thesubclinical progeny, a survival curve was not formulated.

Growth rate differences among progeny groupsRecorded weights of fingerlings at the beginning (time=0) and

end (time=12 weeks) of the experiment were used to calculate growthof fingerlings (g/fish) among the treatment levels for both progenygroups, and the effects of the experimental factors (i.e.-densityand feeding rate) were assessed for both progeny groups. Growthrates for all experimental groups ranged from 6.6 to 10.3 g/fish(Tables 3a and 3b). For the subclinical progeny, no relationshipbetween growth and the experimental factors was evident (p=0.3445;F=1.11; df=2/24); growth rates of experimental groups ranged from7.9 to 9.7 g/fish and did not show consistent correlation witheither density or feeding rate. For the clinical progeny group, aninteraction was evident such that higher growth rates occurred inthe treatment levels with higher density and feeding rates

-44-

(p=0.0545; F=4.10; df=1/23) (Figure 3). A wider range of growthrate was evident for clinical progeny experimental groups (6.6 -10.3 g/fish).

Antigen prevalence by ELISA and DFAT: differences among treatmentlevels of progeny groups

The proportion of ELISA positive fingerlings among alltreatment levels of the subclinical progeny group approached zerofor the 4,8,and 12 week samples (< 0.035)(Table 4a). No furtherassessment of antigen prevalence in relationship to treatment levelwas warranted.

For the clinical progeny group, no relationship was apparentbetween antigen prevalence determined by ELISA and either densitylevel or feeding rate. The proportion of sample fish testingpositive by ELISA for each experimental group at each sampling timeranged from 0.60 to 0.93, with no evident correlation with densityalone, feeding rate alone, or a combination of the two factors(p=0.842, X2=0.0832, df=3) (Table 4b). A wide range in ELISA ODvalues was found for positive samples within each experimentalgroup (see above).

Direct fluorescent antibody testing for the clinical progenygroup was also used to assess antigen prevalence among theexperimental groups of the clinical progeny. Again, no correlationwas evident between the antigen prevalence and the density and/orfeeding rate variables (p=0.276, X2=2.575, df=2)(Table 5).

Measure of Agreement between ELISA and DFATKappa values obtained for each sampling interval (time= 0, 4,

8, and 12 weeks) all indicated a low correlation between ELISA andDFAT results for the clinical progeny group. The Kappa statisticswere calculated to be 0.20 at the 0 week sample, 0.30 at the 4 weeksample, 0.24 at the 8 week sample, and 0.06 at the 12 week sample(with Kappa being equal to 1.0 with perfect correlation and equalto 0 when agreement equals that expected by chance). In contrast,a comparison of agreement between ELISA and DFAT results formortalities from the clinical progeny group yielded a Kappa valueof 1.0, representing perfect correlation (Figure 4).

Covariancy of water quality variables (dissolved oxygen andunionized ammonia concentrations)

Mean dissolved oxygen value ranged from approximately 4.0ppm to 11.0 ppm among experimental groups with the higher valuesassociated with the lower densities and, to a lesser degree,feeding rates. Mean unionized ammonia values, ranging fromapproximately 0.0001 ppm to 0.0007 ppm, were higher in experimentalgroups with the higher densities and feeding rates (Figures 5a and

-45-

6a). Unlike dissolved oxygen values, the unionized ammonia valuesencountered in this investigation generally would not beconsidered detrimental to fish (Dupree and Huner, 1984)(Figures 5band 6b).

Correlation between these two water quality parameters andexperimental groups was evaluated to determine covariancy (i.e.-the extent to which the treatment level and the water qualityparameter vary together/interrelate). For the subclinical progenygroup, significant correlation was evident between density leveland dissolved oxygen content (p=0.0001, F=222.53, df=1/24) as wellas between density level and unionized ammonia level (p=0.0001, F=129.61, df=1/24). Correlation was not found between feeding rateand either of the water quality variables.

For the clinical progeny group, a significant correlation wasagain evident between the density level and the dissolved oxygencontent (p=0.001, F=37.84, df=1/24), but no correlation existedbetween feeding rate and dissolved oxygen. Both feeding rate anddensity level correlated significantly with unionized ammoniacontent for this group (density: p=0.001, F=24.51, df=1/24; feedingrate: p= 00546, F=4.08, df=1/24).

IV. DiscussionThe ELISA results for overall antigen prevalence differences

between the two progeny groups were consistent with previousreports in the literature regarding vertical transmission of R.salmoninarum . Brood stock segregation of chinook salmon based onserological indicators of antigen prevalence (ELISA and DFAT) hasbeen demonstrated to significantly reduce both antigen prevalence(Pascho et al., 1991) and mortality (Elliott et al., 1995) in theoffspring. In this study, the apparent increases in R. salmoninarumantigen prevalence as determined by ELISA and mortality levelassociated with the clinical progeny group over the subclinicalprogeny group lend further support to the importance of brood stocksegregation as an effective means of BKD control among culturedsalmonids. Interpretation of ELISA results based upon the ninetreatment levels for each group was, however, less conclusive. Noapparent correlation between stocking density and/or feeding rateand the prevalence of R. salmoninarum antigen was demonstrated inthis study.

Correlation of the secondary variables (dissolved oxygen andunionized ammonia content of water) with treatment levels ofdensity and feeding rate was evident in this investigation. Themost consistent correlation was observed between the density leveland the two water quality parameters; namely, increased stockingdensity coincided with elevated levels of unionized ammonia anddecreased levels of dissolved oxygen for both progeny groups.

-46-

Water quality variables, particularly the dissolved oxygen contentsbelow 6.0 ppm which occurred in the higher density treatmentgroups, represent potential stress factors directly associated withdensity and feeding rate (Dupree and Huner, 1984); however, stresswas not manifested through increased antigen prevalence ordecreased growth/survival in response to these variables.

The relationship, if any, between excessive levels of stockingdensity and feeding rate and increased R. salmoninarum antigenprevalence among chinook salmon fingerlings was not substantiatedby this investigation. The principle of environmental stressimpacting resistance to disease and antigen prevalence is wellestablished (Snieszko, 1974). While this relationship, as itrelates to BKD pathogenesis, was not conclusively supported by thisstudy, it certainly cannot be dismissed. As no parameter reflectingstress levels (i.e. endogenous serum cortisol levels) was assessedin this investigation, stress differences among the treatmentgroups were not definitively established. Mazur et al. (1993)conducted a similar study examining the effects of stocking densityand feeding rate upon R. salmoninarum antigen prevalence in chinooksalmon smolts held in seacages, obtaining significantly differentresults; that study reported significant density effects uponantigen prevalence such that higher stocking density correlatedwith increased antigen prevalence. Differences in these and ourfindings may be explained by a number of factors related toexperimental design, including life stage of the fish (smoltsversus fingerlings), the habitat (freshwater enclosed tanks versuscages in saltwater), and antigen prevalence assessment (DFAT andELISA versus the quantitative fluorescent antibody technique).Additionally, the Mazur et al. (1993) investigation utilizedtreatment level densities below their standard density whereas ourstudy used treatment level densities at and above a designatedstandard (24 g/l). Our investigation attempted a laboratorysimulation of salmonid culture conditions rather than an experimentconducted on site at a salmonid culture facility. It is quitepossible that given other environmental conditions (water flow,other water parameters, alternative treatment levels for potentialstressors), the outcome may have been substantially different.

Otherwise, the level of infection represented by the twoprogeny groups may have been inappropriate for illustration of theimpacts of environmental parameters on BKD pathogenesis and antigenprevalence. For instance, the infection level of the subclinicalprogeny group may have been too low in the majority of fish fordisease progression to be evident, even with the impact of externalstressors. Likewise, individuals in the clinical progeny group mayhave been so highly infected with R. salmoninarum that any stressand related immunosuppressive facilitation of disease pathogenesiswas not a factor.

-47-

Alternatively, the possibility exists that the findingscorrectly support the notion that environmental variables play aminimal role in BKD pathogenesis of vertically infected juvenilesalmonids. As vertical transmission involves intra-ovum infectionby R. salmoninarum prior to the development of protective immunityin the offspring, immunological tolerance may be a factor inpathogen dissemination and subsequent disease outbreak. Brown etal. (1996) showed that exposure of coho salmon ( Oncorhynchuskisutch ) as unfertilized eggs to the p57 antigen of R. salmoninarumdiminished the subsequent ability of these fish to mount a humoralimmune response against p57. With immunological tolerance for R.salmoninarum , external factors which would typically diminishimmunological function would not influence BKD pathogenesis asfully as would be expected for immunologically competent fish.Different results would therefore be expected with fish infectedhorizontally. Mazur et al. (1993) also discusses the probabilitythat horizontal (fecal-oral) transmission occurred among fish inthat investigation.

Both serological tests employed in this study, ELISA and DFAT,yielded dissimilar results. While the test methodologies display100% correlation in confirming mortality due to BKD, agreementbetween the tests applied to asymptomatic, clinically diseased, orhealthy fish was poor. Both DFAT and ELISA are highly utilized andconsidered reliable diagnostic methodologies common to bothclinical medicine and scientific research. Factors related to eachof these tests specifically adapted for detection of R.salmoninarum may, at least in part, explain the inconsistencybetween results. The ELISA detects soluble antigen of R.salmoninarum that is extracted from tissue, while the DFAT detectswhole cells (i.e. cell surface antigen) of R. salmoninarum . Theoverall ability to detect soluble antigen in homogenized tissuesamples would presumably be greater than the ability to detectcells from a tissue smear representing a comparably far smallertissue sample. This difference has been cited as a contributoryfactor to the overall greater sensitivity of ELISA as compared toDFAT (Meyers et al., 1993; Dixon, 1987). On the other hand,inherent features of this ELISA for R. salmoninarum antigen mayalso undermine its reliability. The use of a polyclonal antiserumto R. salmoninarum may allow for cross-reactivity with similarantigens, producing false positive results; likewise, cross-reactivity with other bacterial species has been reported for theDFAT (Brown et al., 1995; Wood et al., 1995; Austin et al., 1985).Additionally, the complexity of the ELISA (in terms of equipment,reagents, and procedure) compared to DFAT allows for greater errorin technique and interpretation of results. There is considerableprecedence for discrepancies of results among serological tests,ELISA and DFAT in particular, in the literature. Indeed, Meyers etal.(1993) describes considerable dissimilarities in results

-48-

obtained from ELISA and DFAT as methods of assessing prevalence ofR. salmoninarum among salmonid stocks. Complete agreement betweenELISA and DFAT applied to the mortality specimens in this currentstudy suggests that levels of bacterial cells and soluble antigenin fish with lethal BKD are sufficiently elevated to be readilydetected serologically by either method. The otherwise poorcorrelation between the techniques suggests that levels of R.salmoninarum organisms and soluble antigen in subclinically/earlyclinically diseased fish do not necessarily coincide.

In this study, treatment groups with a combination of higherdensity and feeding rates were associated with better survival andgrowth in the clinical progeny group. Assuming that the higherlevels of density and feeding rate are valid representations ofelevated environmental stress, this is an unusual finding; there isnot substantial precedence for physiological benefit from externalstressors in fish culture. Previous literature generally has shownan inverse correlation or lack of correlation between variousenvironmental stressors and survival/growth of fish (Kjartansson etal., 1988; Fagerlund et al., 1981). One possible explanation isthat this investigation was confounded by the presence of increasedlevels of organic debris, including excessive uneaten feed andfecal matter, in the treatment level tanks with the highercombinations of density and feeding rate. As an obligate pathogen,R. salmoninarum has limited survival capabilities outside the hostrelated to an affinity for organic debris, including fecal matter(Austin and Rayment, 1985; Mitchum and Sherman, 1981). It ispossible that horizontal transmission of R. salmoninarum wasdecreased in experimental groups with increased organic debris(i.e.- the higher treatment levels) since the debris provided analternative foci of bacterial colonization. As all experimentaltanks were cleaned generally every other day, the bacteria wouldhave been routinely removed from the system in these treatmentgroups. The overall antigen prevalence data for the clinicalprogeny group (Figure 1) reflects an increase in the proportion ofELISA positive fish throughout the course of the experiment,possibly indicating a component of horizontal transmission.Alternatively, individual fingerlings could simply have beenexperiencing increased antigen loads with time, increasing theproportion of fish with antigen levels surpassing the negativethreshold for ELISA. Furthermore, this hypothesis was notsupported by results for antigen prevalence among the treatmentlevels (i.e.- antigen prevalence did not decrease in highertreatment levels) as determined by either ELISA or DFAT. Furtherinvestigation into the bacterial load in organic debris from thedifferent treatment levels would be warranted to better elucidatethe potential impact of organic debris upon survival and growth ofcultured salmonids as well as upon R. salmoninarum antigenprevalence in tissues.

-49-

Table 1. Organizational scheme for experimental groups.

Organization of the nine experimental groups for each progenypopulation, representing three treatment levels each of stockingdensity (D1-D3) and feeding rate (F1-F3). The lowest levels ofeach (D1/F1; 24 g fish/l and 2.8 % body weight/day) represent"ideal" levels while the other levels represent degrees ofovercrowding and overfeeding.

Stocking density Feeding rate (% body weight/day) (g fish/l)

2.8 % 5.6 % 8.4 %

24 g/l D1/F1 D1/F2 D1/F3

40 g/l D2/F1 D2/F2 D2/F3

48 g/l D3/F1 D3/F2 D3/F3

-50-

Table 2. BKD-related mortalities from the clinically infected progeny group

BKD-related mortalities, tallied on a weekly basis, for each of thenine experimental groups (D1-3 = density levels, D1 = 24 g/l, D2 =40 g/l. D3 = 48 g/l; F1-3 = feeding rates, F1 = 2.8% bodyweight/day, F2 = 5.6 %, F3 = 8.4%) among the clinically diseasedpopulation of chinook salmon fingerlings. The total cumulativenumber of BKD-related mortalities was 1, 218 fingerlings.

Group Total Time (week)

1 2 3 4 5 6 7 8 9 10 11 12

D1F1 26 18 11 8 12 8 4 5 6 3 2 1 104

D1F2 27 13 10 11 13 11 3 5 5 2 0 3 103

D1F3 35 22 16 3 7 5 5 7 1 4 3 4 112

D2F1 40 32 14 18 16 5 5 5 5 2 4 4 150

D2F2 43 34 19 15 11 7 6 6 10 5 5 6 167

D2F3 27 17 14 13 14 9 4 13 7 7 4 7 136

D3F1 28 18 15 17 13 9 6 11 4 7 5 4 137

D3F2 33 15 16 23 25 13 8 5 12 8 5 6 169

D3F3 22 25 16 14 12 8 6 5 9 3 8 12 140

Total: 281 194 131 122 124 75 47 62 59 41 36 47 1,218

-51-

Table 3a. Growth among the asymptomatic progeny population.

Mean growth (g/fish) for the experimental groups of theasymptomatic progeny population over twelve weeks. D1-3 representthe density levels while F1-3 represent the feeding rates.

Stocking Density Feeding rate (F; % body weight/day) (D: g fish/l)

F1 (2.8%) F2 (5.6%) F3 (8.4%)

D1 (24 g/l) 8.0 9.7 7.9

D2 (40 g/l) 8.6 9.1 8.1

D3 (48 g/l) 9.3 9.7 8.4

Table 3b. Growth among the clinical progeny population.

Mean growth (g/fish) for the experimental groups of the clinicalprogeny population over 12 weeks. D1-3 represent the densitylevels while F1-3 represent the feeding rates.

Stocking density Feeding rate (F; % body weight/day) (D; g fish/l)

F1 (2.8%) F2 (5.6%) F3 (8.4%)

D1 (24 g/l) 8.3 8.9 9.1

D2 (40 g/l) 7.9 8.3 10.3

D3 (48 g/l) 6.6 9.6 9.5

-52-

Table 4a. ELISA results for the asymptomatic progeny group.

ELISA data summary for the nine experimental groups at 4, 8, and 12weeks and the time 0 control sample for the asymptomatic progenypopulation. Values represent the proportion of fish with positiveELISA results within each group. D1-3 represent the density levels(D1 = 24 g/l, D2 = 40 g/l, D3 = 48 g/l) and F1-3 represent thefeeding rates (F1 = 2.8% body weight/day, F2 = 5.6%, F3 = 8.4%).

Time (week) Group (D,F) Proportion

0 --- 0.09

4

D1F1 0.02

D1F2 0

D1F3 0.04

D2F1 0.09

D2F2 0.04

D2F3 0.04

D3F1 0

D3F2 0.07

D3F3 0

8

D1F1 0

D1F2 0

D1F3 0

D2F1 0

D2F2 0

D2F3 0.07

D3F1 0.02

D3F2 0

D3F3 0

12

D1F1 0.10

D1F2 0

D1F3 0

D2F1 0

D2F2 0

D2F3 0

D3F1 0

D3F2 0

D3F3 0

-53-

Table 4b. ELISA results for the clinical progeny group.

ELISA data summary for the nine experimental groups at 4, 8, and 12weeks and the time 0 control sample for the clinical progenypopulation. Values represent the proportion of fish with positiveELISA results within each group. D1-3 represent the density levels(D1 = 24 g/l, D2 = 40 g/l, D3 = 48 g/l) and F1-3 represent thefeeding rates (F1 = 2.8% body weight/day, F2 = 5.6 %, F3 = 8.4%).

Time (week) Group (D,F) Proportion

0 ---- 0.50

4

D1F1 0.67

D1F2 0.60

D1F3 0.80

D2F1 0.63

D2F2 0.83

D2F3 0.80

D3F1 0.83

D3F2 0.63

D3F3 0.77

8

D1F1 0.93

D1F2 0.84

D1F3 0.70

D2F1 0.77

D2F2 0.71

D2F3 0.63

D3F1 0.63

D3F2 0.70

D3F3 0.60

12

D1F1 0.76

D1F2 0.71

D1F3 0.73

D2F1 0.73

D2F2 0.79

D2F3 0.77

D3F1 0.81

D3F2 0.71

D3F3 0.77

-54-

Table 5. DFAT results for the clinical progeny group.DFAT data summary for the experimental groups at 4, 8, and 12 weeksand the time 0 control sample from the clinical progeny population.Values represent the proportion of fish within each DFATclassification (0 = negative, 1 = 1-5 cells per field, 2 = greaterthan 5 cells per field) for each experimental group. D1-3represent the density levels (D1 = 24 g/l, D2 = 40 g/l, D3 = 48g/l) and F1-3 represent the feeding rates (F1 = 2.8% bodyweight/day, F2 = 5.6%, F3 = 8.4%).

Time Group (week) (D,F)

Proportion per DFAT category

0 1 2

0 --- 0.12 0.56 0.32

4

D1F1 0.50 0.33 0.17

D1F2 0.56 0.27 0.17

D1F3 0.42 0.27 0.31

D2F1 0.40 0.37 0.23

D2F2 0.43 0.30 0.27

D2F3 0.50 0.42 0.08

D3F1 0.37 0.37 0.26

D3F2 0.59 0.33 0.08

D3F3 0.12 0.60 0.28

8

D1F1 0 0.60 0.40

D1F2 0.20 0.52 0.28

D1F3 0 0.80 0.20

D2F1 0.08 0.56 0.36

D2F2 0.28 0.58 0.14

D2F3 0.18 0.62 0.20

D3F1 0.20 0.60 0.20

D3F2 0.10 0.60 0.30

D3F3 0.12 0.38 0.60

12

D1F1 0.65 0.30 0.05

D1F2 0.44 0.38 0.18

D1F3 0.58 0.32 0.10

D2F1 0.60 0.28 0.12

D2F2 0.60 0.30 0.10

D2F3 0.70 0.27 0.03

D3F1 0.54 0.42 0.04

D3F2 0.70 0.28 0.02

D3F3 0.50 0.40 0.10

-55-

Figure 1: A comparison of R. salmoninarum antigen prevalence as determined by ELISA for two progeny groups.

The y-axis displays the proportion of fish sampled within each population which tested positive for the antigen.

-56-

Figure 2: Survival analyses for the nine treatment groups of the clinical progeny population.(ch1fig2.pic; 828K)

The density variable is represented by the columns (left=24 g/l, middle= 40 g/l, right=48 g/l) and thefeeding rate variable is represented by the rows (top=8.4%, middle=5.6%, bottom=2.8%). Note thehighest survival rates at the end of the 12 week interval were experienced by the groups with thehigher levels of density and feeding (survival curves furthest top and right).

-57-

Figure 3: Growth rates for the treatment groups from the clinicalprogeny population.

Each bar color represents a different density levelat each feeding rate. Note the increase in growth rate corresponding to increased density for each feeding rate, especially apparent for the 5.6% and 8.4% feeding rates.

-58-

Figure 4:Correlation between the ELISA and DFAT results

for the clinical progeny population. (Perfect correlation is represented by a Kappa value of 1.00 and absence ofcorrelation is represented by a Kappa value of zero.)

-59-

Figure 5: Covariancy of water quality parameters, dissolved oxygen(5a) and unionized ammonia (5b), with the treatment groups for the subclinical progeny population.

Each bar color represents a different density level at each feeding rate.

-60-

Figure 6: Covariancy of water quality parameters, dissolvedoxygen (6a) and unionized amonia (6b), with thetreatment groups for the clinical progeny population. Each bar color represents a different density level ateach feeding rate.

-61-

References:

Agresti, A. 1990. Categorical Data Analysis. John Wiley andSons, New York, NY. 558 p.

Austin, B., D. Bucke, S. Feist, and J. Rayment. 1985. A falsepositive reaction in indirect fluorescent antibody test forRenibacterium salmoninarum with a coryneform organism. Bulletinof European Association of Fish Pathologists. 5:8-9.

Austin, B. and J. Rayment. 1985. Epizootiology ofRenibacterium salmoninarum , the causal agent of bacterial kidneydisease in salmonid fish. Journal of Fish Diseases. 8:505-509.

Bowser, P.R., R.B. Landy, and G.A. Wooster. 1988. Efficacy ofelevated dietary fluoride for the control of Renibacteriumsalmoninarum infection in rainbow trout, Salmo gairdneri . Journal of the World Aquaculture Society. 19(1):1-7.

Bruno, D.W. and A.L.S. Munro. 1986. Observations onRenibacterium salmoninarum and the salmonid egg. Diseases ofAquatic Organisms. 1:83-87.

Brown, L., G.K. Iwama, and T.P.T. Evelyn. 1996. The effect ofearly exposure of coho salmon eggs to the p57 protein ofRenibacterium salmoninarum on the development of immunity to thepathogen. Fish & Shellfish Immunology 6: 149-165.

Brown, L.L., T.P.T. Evelyn, G.K. Iwama, W.S. Nelson, and R.P.Levine. 1995. Bacterial species other than Renibacteriumsalmoninarum cross-react with antisera against R. salmoninarumbut are negative for the p57 gene of R. salmoninarum detected bythe polymerase chain reaction (PCR). Diseases of AquaticOrganisms. 21:227-231.

Bullock, G.L., B.R. Griffin, and H.M. Stuckey. 1980. Detection of Corynebacterium salmoninus by direct fluorescentantibody test. Canadian Journal of Fisheries and AquaticScience. 37:719-721.

Dixon, P.F. 1987. Comparison of serological techniques forthe identification of Renibacterium salmoninarum . Journal ofApplied Ichthyology. 3:131-138.

Dupree, H.K. and Huner, J.V. 1984. Third Report to the FishFarmers: The Status of Warmwater Fish Farming and Progress inFish Farming Research. U.S. Fish and Wildlife Service, Washington, D.C., USA. 270 p.

-62-

Elliott, D.G., R.J. Pascho, and A.N. Palmisano. 1995. Broodstock segregation for the control of bacterial kidney disease canaffect mortality of progeny chinook salmon ( Oncorhynchustshawytscha ). Aquaculture. 132:133-144.

Elliott, D.G., R.J. Pascho, and G.L. Bullock. 1989. Developments in the control of bacterial kidney disease ofsalmonid fishes. Diseases of Aquatic Organisms. 6:201-213.

Evelyn, T.P.T., L. Prosperi-Porta, and J.E. Ketcheson. 1986. Persistence of the kidney disease bacterium, Renibacteriumsalmoninarum , in coho salmon, Oncorhynchus kisutch (Walbaum),eggs treated during and after water-hardening with povidone-iodine. Journal of Fish Diseases. 9:461-464.

Evendon, A.J., T.H. Grayson, M.L. Gilpin, and C.B. Munn. 1993. Renibacterium salmoninarum and bacterial kidney disease - theunfinished jigsaw. Annual Review of Fish Diseases. 1993:87-104.

Fagerlund, U.H.M., J.R. McBride, and E.T. Stone. 1981. Stress-related effects of hatchery rearing density on cohosalmon. Transactions of the American Fisheries Society. 110:644-649.

Kjartansson, H., S. Fivelstad, J.M. Thomassen, and M.J. Smith. 1988. Effects of different stocking densities on physiologicalparameters and growth of adult Atlantic salmon ( Salmo salar L.)reared in circular tanks. Aquaculture. 73:261-274.

Laidley, C.W. and J.F. Leatherland. 1988. Cohort sampling,anaesthesia, and stocking density effects on plasma cortisol,thyroid hormone, metabolite, and iron levels in rainbow trout,Salmo gairdneri Richardson. Journal of Fish Biology. 33:73-88.

Lall, S.P., W.D. Paterson, J.A. Hines, and N.J. Adams. 1985. Control of bacterial kidney disease in Atlantic salmon, Salmosalar L., by dietary modification. Journal of Fish Diseases. 8:113-124.

Leatherland, J.F. and C.Y. Cho. 1985. Effect of rearingdensity on thyroid and interrenal gland activity and plasma andhepatic metabolite levels in rainbow trout, Salmo gairdneriRichardson. Journal of Fish Biology. 27:583-592.

Mazur, C.F., D.Tillapaugh, and G.K. Iwama. 1993. The effects offeeding level and rearing density on the prevalence ofRenibacterium salmoninarum in chinook salmon ( Oncorhynchustshawytscha ) reared in salt water. Aquaculture. 117:141-147.

-63-

Meyers, T.R., S. Short, C. Farrington, K. Lipson, H.J. Geiger,and R. Gates. 1993. Comparison of the enzyme-linkedimmunosorbent assay (ELISA) and the fluorescent antibody test(FAT) for measuring the prevalences and levels of Renibacteriumsalmoninarum in wild and hatchery stocks of salmonid fishes inAlaska, USA. Diseases of Aquatic Organisms. 16:181-189.

Mitchum D.L. and L.E. Sherman. 1981. Transmission ofbacterial kidney disease from wild to stocked hatchery trout. Canadian Journal of Fisheries and Aquatic Science. 38:547-551.

Pascho, R.J.,D.G. Elliot, and J.M. Streufert. 1991. Broodstock segregation of spring chinook salmon Oncorhynchustshawytscha by use of the enzyme-linked immunosorbent assay(ELISA) and the fluorescent antibody technique (FAT) affects theprevalence and levels of Renibacterium salmoninarum infection inprogeny. Diseases of Aquatic Organisms. 12:25-40.

Pascho, R.J. and D. Mulcahy. 1987. Enzyme-linkedimmunosorbent assay for a soluble antigen of Renibacteriumsalmoninarum , the causative agent of salmonid bacterial kidneydisease. Canadian Journal of Fisheries and Aquatic Science. 44:183-191.

Pickering, A.D. and A. Stewart. 1984. Acclimation of theinterrenal tissue of the brown trout, Salmo trutta L., to chroniccrowding stress. Journal of Fish Biology. 24:731-740.

Piper, R.G., I.B. McElwain, L.E. Orme, J.P. McCraren, L.G.Fowler, and J.R. Leonard. 1982. Fish Hatchery Management. USDI, Fish and Wildlife Service, Washington, D.C., USA. 517 p.

Sanders, J.E., K.S. Pilcher, and J.L. Fryer. 1978. Relationof water temperature to bacterial kidney disease in coho salmon( Oncorhynchus kisutch ), sockeye salmon ( O. nerka ), and steelheadtrout ( Salmo gairdneri ). Journal of the Fisheries ResearchBoard of Canada. 35:8-11.

Snieszko, S.F. 1974. The effects of environmental stress onoutbreaks of infectious diseases of fishes. Journal of FishBiology. 6:197-208.

Suzimoto, B.K., C.B. Schreck, and J.D. McIntyre. 1977. Relative resistance of three transferrin genotypes of coho salmon( Oncorhynchus kisutch ) and their heamatological responses tobacterial kidney disease. Journal of the Fisheries ResearchBoard of Canada. 34:1-8.

-64-

Wedemeyer, G.A. and W.T. Yasutake. 1977. Clinical methods forthe assessment of the effects of environmental stress on fishhealth. Technical Papers of the U.S. Fish and Wildlife Service89. Washington, D.C. 18 p.

Whalen, K.G. and G.W. LaBar. 1994. Survival and growth ofAtlantic salmon ( Salmo salar ) fry stocked at various densities inthe White River, Vermont. Canadian Journal of Fisheries andAquatic Science. 51:2164-2169.

Wolf, K. and C.E. Dunbar. 1959. Methods of infecting troutwith kidney disease and some effects of temperature onexperimental infections. U.S.D.I., F.W.S. Special ScientificReport - Fisheries No. 286. 8 p.

Wood, P.A., G.D. Wiens, J.S. Rohovec, and D.D. Rockey. 1995. Identification of an immunologically cross-reactive 60-kilodaltonRenibacterium salmoninarum protein distinct from p57:Implications for immunodiagnostics. Journal of Aquatic AnimalHealth 7(2):95-103.

Acknowledgments:

I wish to thank my co-investigators: Dr. Cliff Starliper,Dr. Larisa Ford, Dr. Rocco Cipriano, and Dr. David Smith. Wealso wish to thank the investigators and technical staff of theNational Fish Health Research Laboratory, Kearneysville, WV fortheir assistance with this investigation. In particular, wethank Dr. Jeff Teska for his help with the ELISA protocol andTeresa Shatzer for her assistance with the data collection.

-65-

Chapter 2

In vivo effects of the extracellular protein of Renibacterium salmoninarum upon immunological

parameters in juvenile rainbow trout ( Oncorhynchus mykiss) .

-66-

Abstract

Juvenile rainbow trout ( Oncorhynchus mykiss ) exposed in vivo to theextracellular protein (ECP) of Renibacterium salmoninarum wereevaluated in terms of various immunological parameters, includingthe ability to mount a secondary immune response against anotherpathogen of salmonids. Seven groups of one hundred trout were eachtreated as follows: intraperitoneal (IP) injection of ECP (150 µgECP/fish) and IP vaccination against Yersinia ruckeri with aformalin-killed bacterin (4 groups); vaccination alone; ECPinjection alone; IP injection of phosphate buffered saline as acontrol. For the four groups receiving both the ECP andvaccination, the timing was offset so that the ECP was given: 1 dayprior to vaccination; simultaneously with vaccination; 8 days post-vaccination; or 22 days post-vaccination. Twenty-two days post-vaccination, all groups were challenged with Y. ruckeri . At 12sampling times, pre and post-challenge, the following immunologicalparameters were evaluated in seven fish from each of the sevengroups: hematocrit, total serum protein, total leukocyte count,differential leukocyte count, microagglutination titer, and splenicphagocytosis. Post-challenge mortalities were recorded andcultured for Y. ruckeri . Results of the investigation indicatedimmunomodulatory effects of the ECP upon some of the selectparameters. The ECP-exposed fish showed post-challenge elevationsin leukocyte counts and small lymphocyte counts, representing apotentially immunostimulatory effect. The ECP exposure precludeda post-challenge increase in monocyte numbers, representing apotentially immunosuppressive effect of the ECP. Mortality relatedto Y. ruckeri challenge occurred only in unvaccinated fish, withfewer mortalities in the group receiving the ECP alone than in thecontrol group, possibly indicating ECP-related nonspecificimmunostimulation. No effects related to ECP exposure were evidentfor the other parameters examined.

-67-

I. IntroductionBacterial kidney disease (BKD) of salmonid fish is a chronic,

granulomatous disease often associated with high levels ofmorbidity and mortality. The causative agent, Renibacteriumsalmoninarum , is a Gram-positive diplobacillus. This organism hasa number of unique features involved in disease pathogenesis andimmunomodulation. The extracellular protein (ECP) produced by R.salmoninarum contains a number of soluble antigens. The ECP may beisolated from both bacteria in artificial media and from BKD-infected salmonid tissues at concentrations approaching 1 mg/ml(Turaga et al., 1987a; Kaattari et al., 1989), and it has beenextensively investigated as a factor in BKD pathogenesis.Antigenically, the most prominent component of the ECP is a 57 kDprotein (p57) deemed the major soluble antigen or Antigen F(Getchell et al., 1985). Other antigenic ECP constituents, withmolecular weights both higher and lower than the 57 kD antigen,have been identified by a number of investigators (Getchell et al.,1985; Fiedler and Draxl, 1986; Wiens and Kaattari, 1989; Griffithsand Lynch, 1991; Bandin et al., 1992). A number of R. salmoninarumvirulence factors have been attributed to the ECP, particularly tothe p57 antigen. Included in these are the hydrophobicity andautoaggregation of R. salmoninarum cells, agglutination of rabbiterythrocytes, and agglutination of salmonid spermatozoa (Daly andStevenson, 1987; Daly and Stevenson, 1989). The p57 protein hasalso been associated with cellular adhesion by Dubreuil et al.(1990) who identified this antigen to be a cell surface fimbriae.

Various immunomodulatory effects have also been ascribed tothe ECP, particularly to the p57 antigen. In vitroleukoagglutination of salmonid leukocytes from the pronephros,spleen, and peripheral circulation, and induction of macrophages invitro are among the effects associated with the p57 antigen of R.salmoninarum (Kaattari et al., 1989). Similarly, Turaga et al.(1987b) described immunomodulation associated with soluble antigensof R. salmoninarum , characterized as an in vitro decrease inantibody responses to an unrelated antigen. Otherimmunomodulatory mechanisms have been associated with BKD withoutregard to any association with soluble antigens; these include theintracellular survival and reproduction of R. salmoninarum (Bandinet al., 1993) and the involvement of immunological hypersensitivityin BKD pathogenesis (Sami et al., 1992; Evendon et al., 1993).

The pervasion of any in vivo immunomodulation by R.salmoninarum and the importance of these immunomodulatory featuresin epizootics of BKD or other diseases is largely undetermined.Moreover, the general in vivo effects of the soluble antigens of R.salmoninarum have not been fully elucidated. Shieh (1988)identified a lethal extracellular toxin produced in vivo by R.salmoninarum . In contrast, Bandin et al. (1991) concluded that no

-68-

in vivo toxic effects associated with the ECP of R. salmoninarumwere evident.

Immunomodulatory mechanisms enacted by R. salmoninarum uponthe host may be reflected by the host's ability to mount aneffective immune response to another pathogen following exposure tothe immunomodulatory agent. The objective of this investigationwas to examine the in vivo immunomodulatory effects of the R.salmoninarum ECP in salmonid fish as they influence the ability ofthe salmonid host to mount an immune response against anotherlethal pathogen.

II. Materials and Methods

Preparation of the Extracellular Protein Extracellular protein was recovered by a modification of the

methods of Hsu et al. (1991) and Rockey et al. (1991). An isolateof the ATCC #33209 strain of R. salmoninarum was subcultured andincubated on KDM2 agar at 15 C for 21 days. Samples from theo

cultures were Gram-stained and serologically evaluated via thedirect fluorescent antibody test (DFAT) for R. salmoninarum(Bullock et al., 1980) in order to confirm pure cultures of theorganism. The plates were rinsed with sterile water, and thebacterial colonies agitated to remove them from the agar. Thebacterial suspension was centrifuged for 20 min. at 600 xg and10 C. The supernatant was recovered, and protein concentration waso

determined by the method of Bradford (1976). Sterile phosphatebuffered saline (PBS; pH=7.2) was used to dilute the solution tothe final concentration of 750 µg/ml. The extracellular proteinsolution was maintained at 15 C for 24 h. preceding injection intoo

sample fish.

Preparation of Yersinia ruckeri bacterinA formalin-killed particulate Yersinia ruckeri (causative

agent of enteric redmouth disease, ERM, of salmonids) antigenpreparation was developed by a modification of the method ofAnderson and Dixon (1989). Briefly, the 11.40 isolate of Serotype1 Y. ruckeri (Schill et al., 1984) was subcultured in tryptic soybroth medium. Bacterial cells were harvested from log phase growthvia centrifugation (150 xg for 15 min.), and the pelleted cellswashed three times in PBS. Cells were resuspended in PBS to a 1%solution, and formalin (0.4% final volume) was added. Thesuspension was continuously stirred for 24 h. and stored at 4 C. o

Prior to vaccination, the suspension was readjusted with PBS to aconcentration yielding 30% transmission at 525 nm on aspectrophotometer.

-69-

Preparation of Y. ruckeri bath suspensionAn isolate of Y. ruckeri (11.40) was cultured on tryptic soy

agar, suspended in sterile PBS, and injected into ten juvenilerainbow trout to confirm viability and pathogenicity. Bacterialcultures from renal tissues were taken from the resultingmortalities, and Y. ruckeri was re-isolated and biochemicallyidentified. Eight liters of tryptic soy broth were inoculatedwith the pathogenic isolate. After 48 h. incubation at 22 C, O

bacterial cells were precipitated via centrifugation (6000 xg for15 min.) and resuspended in PBS to a total volume of 1400 ml. Cellconcentration was quantified via dilution plate counts (Cipriano etal., 1991), and the suspension was divided into seven 200 mlaliquots. One aliquot was added to each experimental tank at thetime of challenge.

Experimental groups and sampling chronologyOne hundred juvenile rainbow trout ( Oncorhynchus mykiss ;

average weight = 65 g) were arbitrarily distributed into each ofseven circular 285 l tanks supplied with spring water (12.5 C). o

Fish were maintained on a commercial diet of pelleted salmonid feed(Zeigler Brothers, Inc., Gardners, PA). Each group was treatedwith an experimental regime as follows: vaccination with the Y.ruckeri bacterin (0.2 ml IP); injection with the R. salmoninarumECP solution (0.2 ml IP; 150 µg ECP/fish); injection with thebacterin and the ECP (4 groups; 0.2 ml IP each solution); andinjection with sterile PBS (0.2 ml IP). For the four groupsinjected with both bacterin and ECP, the chronology of theinjections was varied as follows: simultaneous injection withbacterin and ECP; injection of ECP one day prior to vaccination;injection with ECP 8 days following vaccination; injection with ECPat time of challenge (22 days following vaccination). Seven fishwere removed from the appropriate groups for sampling (as describedbelow) at the following times post-ECP injection/vaccination: 12hours, 1 day, 2 days, 4 days, 8 days, 16 days, and 22 days. For thetwo groups in which the vaccination and ECP injection were offsetby 8 and 22 days, sampling was done at the aforementioned timeintervals post-ECP injection. Fish were sacrificed with a lethaldose of tricaine methanesulfonate (MS-222; Sigma Chemical Co., St.Louis, MO).

Twenty-two days following vaccination (after the 22 day samplewas collected), all groups of fish were challenged with Y. ruckeriin an 80 l bath suspension of 6.5x10 organisms/l for 20 minutes.10

Post-challenge sampling was resumed for all experimental groups asbefore with additional samplings at: 23 days (1 day post-challenge), 26 days (4 days post-challenge), 30 days (8 days post-challenge), 38 days (16 days post-challenge), and 52 days (30 dayspost-challenge). All mortalities were recorded and cultured forY. ruckeri .

-70-

Immunological sampling methodologiesHematocrit/serum protein determination:Whole blood was obtained via veinupuncture of the caudal

vessels using heparinized tuberculin syringes with 22 gaugeneedles. The blood was transferred to capillary tubes andcentrifuged in a microhematocrit centrifuge for 3 minutes.Hematocrit values were assessed using a Critocap reader (SherwoodMedical, St. Louis, MO), and serum protein was measured with aprotein refractometer (Aloe-Hitachi, Japan).

Total leukocyte count:Ten microliters of whole blood from the veinupuncture was

dispensed into a solution of 790 µl PBS with 200 µl Wright's stain(1:100 dilution), and the cells were gently agitated to provideeven distribution. Leukocytes were counted using a hemacytometer,and the counts were extrapolated to obtain the white blood cellconcentration in peripheral blood (Klontz, 1994).

Leukocyte differential count:Ten microliters of whole blood from each fish was spread

evenly across a microscope slide and allowed to air dry. Slideswere methanol-fixed and stained with Leukostat staining solutions(Fisher Diagnostics, Pittsburgh, PA). One hundred leukocytes fromeach sample were categorized cytologically according to type: smalllymphocyte, large lymphocyte, polymorphonuclear cell (PMN),monocyte (Yasutake and Wales, 1983; Campbell, 1988). Thepercentage of each cell type present was multiplied by the totalleukocyte count for the given sample to determine the totaldifferential cell counts.

Microagglutination antibody titer:The remaining blood from each fish was placed into a

Vacutainer (red top) tube (Becton-Dickinson Vacutainer Systems,Rutherford, NJ) and held for six to eight hours at 4 C. The tubes o

were centrifuged at 300 xg for 10 min., and the serum was removedand stored at -70 C. Microagglutination titers against Y. ruckerio

were determined in all samples simultaneously at the completion ofall sampling. Log dilutions of the serum samples were carried out2to twelve dilutions in V-bottom microtiter plates (50 µl/well).Fifty microliters of the Y. ruckeri antigen preparation (30%transmission at 525 nm) previously described were added to eachwell, and microtiter plates were sealed and incubated at 22 C for o

18 hours. Following incubation, the titers were determined,calculated as the reciprocal of the lowest reactive dilution (i.e.the well with the last agglutination reaction). Negative controlwells with PBS were run with the experimental serum samples.

-71-

Phagocytosis assay:Phagocytosis capabilities of splenic phagocytes for Y. ruckeri

were assessed via a modification of the technique described byAnderson (1992a). Briefly, spleens were aseptically removed fromall fish and placed in cold (4 C) sterile Eagle's Minimal Essentialo

Media (EMEM; Earle's salts containing 2% newborn calf serum, 100mg/ml gentomycin, and 0.016 M Tris; Wolf and Quimby, 1973).Spleens were finely macerated, and the cellular debris resuspendedin cold, sterile EMEM. Cell suspensions were washed twice in EMEMvia centrifugation for 5 min. at 90 xg and 4 C. Following theo

second wash, cells were resuspended in 0.51 ml of EMEM. Tenmicroliters of the suspensions were removed for determination ofapproximate leukocyte concentration with a hemacytometer aspreviously described, and the cell suspensions were adjusted to 10million leukocytes/ml. Approximately 5 million leukocytes (0.5 mlsuspension) per sample were placed on sterile microscope slides andincubated at 4 C for 2 hours. Following incubation, the slideso

were gently washed in sterile PBS, and 0.5 ml of the Y. ruckeriantigen preparation (30% transmission at 525nm) diluted 1:10 in PBSwas added to each slide. The slides were again incubated for 2hours at 4 C, gently rinsed in PBS, and allowed to air dry. Onceo

dry, the slides were methanol-fixed and stained with the Leukostatstaining solutions. Stained slides were examined and phagocyticcells were identified based on cellular morphology. One hundredphagocytes were counted per slide, and phagocytosis was assessedbased on the percentage of phagocytic cells which had adhered orengulfed rod-shaped bacteria. Five slides per experimental groupper sampling time were examined. In addition, two slides per groupfrom all post-challenge samples were assessed without the additionof the antigen preparation; these slides were prepared in order todetect any background presence of Y. ruckeri in the infected fish.

Survival:Mortalities were removed daily from all experimental groups

throughout the experiment. Renal tissues from all post-challengemortalities were cultured on eosin/methylene blue agar, and anybacterial colonies were analyzed biochemically to identify themicroorganism(s).

Statistical analysisStatistical analyses were performed with the analysis of

variance (ANOVA) for the data obtained by the following techniques:hematocrit, serum protein, leukocyte count, differential leukocytecount, antibody titer, and phagocytosis. Additionally, for theantibody titer and differential leukocyte count data, the naturallogarithms of the experimental values were substituted into thestatistical calculations to normalize the sample data. For thesurvival data, the Fisher's exact test was used to evaluate

-72-

survival in all groups; further comparison of groups withmortalities was completed using the log-rank statistic. Allstatistical analyses were done with SAS programs (PROCGLM,PROCLIFETEST; SAS Institute, Inc., Cary, NC). Data for thefollowing four groups: PBS-injected, bacterin-injected, ECP-injected, and simultaneous bacterin and ECP-injected, were analyzedas a unit. Data for the four groups injected with ECP and bacterinat different time intervals were also analyzed separately toevaluate any differences based on chronology of the ECP exposure.

III. Results Data is summarized in Tables 1-9 in the appendices of thischapter.

Exposure to the ECP appeared to have no significant effect(p<0.05) on either hematocrit or serum protein values. For bothparameters, however, there was an interaction between vaccinationstatus and challenge with Y. ruckeri , such that both hematocrit andserum protein values decreased significantly post-challenge in theexperimental groups that were not vaccinated. Such decreases werenot noted in the groups which were vaccinated (Figure 1).

Total leukocyte count analysis revealed a significant post-challenge leukocytosis only in the experimental groups exposed tothe ECP (p<0.0062, F=7.59, df=1). Chronological evaluation of theECP effect (i.e.-statistical analysis of the 4 groups vaccinatedand injected with ECP at various times) showed that the leukocytecount decreased as the ECP/bacterin injection interval increased;there was a significant difference between the earliest (time = 1day prior to vaccination) and latest (time = 22 days/at challenge)treatments, but not between the intermediate treatments (p<0.0025,F=4.99, df=3; Figure 2).

Differential leukocyte counts also revealed significanttreatment-based differences in the total numbers of the four majortypes of white blood cells identified. In keeping with the totalleukocyte count results, the small lymphocytes count increasedsignificantly post-challenge in the experimental groups exposed tothe ECP, but not in the unexposed groups (p<0.0139, F=6.11, df=1).Chronological evaluation also showed the small lymphocyte count todecrease with an increasing ECP injection/ vaccination interval,with a significant difference only between the earliest and latestgroups (p<0.0012, F=5.57, df=3). A post-challenge increase insmall lymphocyte count was also evident for the vaccinated fish(p<0.0.04, F=4.25, df=1; Figure 3). For the large lymphocytes, nosignificant changes were evident based upon the exposure to ECP orthe chronology of this exposure. An interactive effect between

-73-

vaccination and challenge was however evident, such that the largelymphocyte count increased significantly post-challenge inunvaccinated fish but not in the vaccinated fish (p<0.0001,F=22.76, df=1; Figure 4). Similarly, there was no apparent effectof ECP exposure upon PMN count, but an interactive effect basedupon vaccination and challenge was again evident. Unvaccinatedfish showed a significantly increased PMN count post-challenge,while the vaccinated counterparts showed a decreased post-challengePMN count (p<0.0001, F=34.72, df=1; Figure 5). Finally, themonocyte count showed a significant increase post-challenge only infish not exposed to ECP (p<0.0246, F=5.10, df=1); no ECP effectbased upon chronology was evident. Additionally, a significantpost-challenge increase in monocyte count was evident inunvaccinated fish but not in the vaccinated fish (p<0.0001,F=16.94, df=1; Figure 6).

The microagglutination antibody titer data revealed nosignificant differences in titers based upon exposure/chronology ofexposure to the ECP. There was, however, a significant elevationin the post-challenge antibody titers among vaccinated fish(p<0.0001, F=32.46, df=1; Figure 7).

Phagocytosis data showed no significant changes based uponeither ECP exposure or vaccination. There was an overallsignificant decrease in the percentage of phagocytes which hadadhered or engulfed bacteria post-challenge compared to pre-challenge (p<0.0014, F=10.40, df=1; Figure 8).

The survival data showed that mortalities in the 30 day post-challenge period occurred only in the groups receiving PBS alone(7 mortalities) and the group receiving the ECP without thebacterin (2 mortalities). Post-mortem renal cultures andbiochemical analyses confirmed the presence of Y. ruckeri in allmortalities. Statistical analysis revealed significantly differentsurvival rates among the seven groups (p<0.0001). Further analysisof the groups with mortalities via survival estimates showed thedifference in survival between the two groups not to bestatistically significant (p<0.05), although survival wasconsiderably lower for the PBS only group than the ECP-exposedgroup (p=0.0799; Figure 9).

IV. DiscussionAs a potential immunomodulatory agent, the R. salmoninarum ECP

has the possibility of effecting not only BKD development but alsoother disease processes in salmonid culture. Given the typicalchronic, slowly progressive nature of BKD, long term presence of R.salmoninarum soluble antigen within the tissues of a host is quitelikely. Thus, this antigen may substantially effect the immunesystem of a salmonid host, altering the immunological response to

-74-

other pathogens. In this respect, subclinical BKD may therefore beregarded as a factor in epizootics of other salmonid diseases. Thequestion remains as to whether any ECP-related immunomodulation isimmunostimulatory, immunosuppressive, or both.

Results of this investigation support the concept ofimmunomodulation brought about by the R. salmoninarum ECP. Theeffect, however, is not consistent among the various immunologicalparameters examined. There appeared to be no ECP-related effectsupon hematocrit, serum protein, antibody microagglutination titer,or phagocytosis capabilities at the 150 µg/fish ECP concentrationused in this investigation. The absence of significant changes inthe hematocrit values indicates no ECP-related changes inerythrocyte production or utilization to a degree whichsubstantially affected the hematocrit. Similarly, no effect wasnoted on serum protein, indicating that total levels of the variousimmunological proteinaceous components of salmonid serum, includingimmunoglobulin, acute phase proteins, lectins, and antibacterialproteinases (Alexander and Ingram, 1992), were unaltered. Lack ofresponse in the antibody titer suggests that ECP does not play aprominent immunomodulatory role in specific humoral immunity invivo . Moreover, the absence of significant alterations in thephagocytic capabilities of the splenic phagocytes suggests thatthis functional aspect of cell-mediated immunity is also notaffected by the ECP in this investigational framework. The lackof effects cannot, however, conclusively demonstrate that theseaforementioned immunological parameters are completely unaffectedin a naturally occurring infection due to experimental limitationsimposed by the use of a single concentration of ECP and uncertaintyregarding in vivo longevity of this ECP following a single dosage.

The total leukocyte count and differential analysisdemonstrated significant changes relating to exposure to the ECP.The increased leukocyte count post-challenge was most likelyrelated directly to the post-challenge increase in small lymphocytenumbers in ECP-exposed fish, as small lymphocytes represented aconsistently large percentage of the total leukocyte count.Interestingly, this relationship appeared strongest (i.e. highestlymphocyte counts) when the vaccination and ECP injection wereperformed close together, with the count diminishing as the ECP wasgiven later (8 and 22 days post-vaccination). These results aredifficult to define in terms of precise immunomodulatory effects assmall lymphocytes were not further differentiated according tofunction. For instance, morphological assessment alone did notdistinguish the Ig+ from Ig- lymphocytes, and this distinction hasconsiderable ramifications for immunological function (Kaattari,1992). It is also uncertain whether the increase in smalllymphocyte numbers in peripheral circulation in this investigationrepresents an increased total cell production or altered leukocyte

-75-

trafficking. It is likely that the elevation in small lymphocytenumbers represents an immunostimulatory response as decreasedlymphocyte counts have been associated with stressors such asexogenous glucocorticoid administration and immunosuppression(Campbell, 1988).

The monocyte count also changed significantly, increasingpost-challenge only in the absence of the ECP. As in mammalianspecies, salmonid monocytes are precursors to macrophages, playingan important role in phagocytosis, cytokine production, andantigen presentation (Secombes and Fletcher, 1992). The results ofthis investigation suggest that the ECP may act as animmunosuppressive agent upon the monocyte population, preventing apost-challenge monocytosis. Indicators of monocyte/macrophagefunction (i.e. phagocytosis assay), however, failed to corroborateadditional immunosuppressive effects for this cell population.

The survival analysis data provides further suggestion ofimmunomodulation by the ECP. The only two groups experiencing Y.ruckeri related mortalities post-exposure were the unvaccinatedgroups injected with PBS or ECP; however, the experimental groupgiven only ECP yielded just two mortalities while the PBS-injectedgroup had seven mortalities in the 30 days post-challenge. Theupper confidence limit for survival of the ECP-injected group isinclusive of 100% (proportion=1.00) survival, experienced by thevaccinated fish; however, it must be noted that a statisticallysignificant difference in survival (p< 0.05) was not observedbetween the ECP-injected and PBS-injected groups. These resultsmay possibly indicate an immunostimulatory effect of the ECP,contributing to an overall enhanced immunological capability andresulting in comparatively better survival in the ECP-injected fishthan in the control group. This notion is not without precedent;Amend and Johnson (1984) similarly found that a pH-lysed R.salmoninarum bacterin was able to potentiate protection againstAeromonas salmonicida when given as a bivalent vaccine along withthe A. salmonicida bacterin. It is interesting to note that otherbacterial preparations, including the lipopolysaccharide of Gram-negative organisms and the killed Mycobacterium incorporated incomplete Freund’s adjuvant have been shown to be immunostimulatoryin piscine species (Anderson, 1992b). The immunostimulatoryeffects of Mycobacterium antigens are particularly intriguing incomparison to the potential for analogous effects with R.salmoninarum antigens, given the similar pathogenesis of theinfections induced by the two agents (i.e. chronic granulomatousdiseases involving facultatively intracellular pathogens andprominent cell-mediated immune responses).

-76-

Regarding the intracellular localization of R. salmoninarum ,it is important to note that this investigation presented ECP tothe host tissues presumably in the extracellular space. It isquite possible that intracellular exposure of salmonid immunocytesto the ECP would produce an entirely different result. Therefore,based on these and previous findings, further investigation intothe possibility of immunological potentiation by the R.salmoninarum ECP is warranted.

Additional findings in this investigation pertaining to theeffects of vaccination upon the immunological parameters alsoproved interesting. Decreased post-challenge hematocrit and serumprotein levels in unvaccinated fish (the groups experiencingmortalities) may be the effects of clinical ERM, a hemorrhagicdisease. Differential leukocyte alterations including theincreases in numbers of large lymphocytes, PMN's, and monocytes inunvaccinated fish may again be a response to ERM and the resultingbacteremia; increased numbers of small lymphocytes post-challengein vaccinated fish may represent an increased level of "memory"cells in the secondary immune response. Again, without thedifferentiation of functional lymphocyte populations, however, thistheory remains speculatory. The post-challenge increase inantibody microagglutination titer in vaccinated fish was expectedas the secondary humoral response against Y. ruckeri antigen insalmonids is well documented (Austin and Austin, 1993). The lackof a vaccination-related stimulation of phagocytosis was also notunexpected as cell-mediated immunity generally is poorly induced bybacterins (Tizard, 1987).

-77-

Figure 1a: Hematocrit. Hematocrit values for vaccinated and unvaccinated fish, pre and post-challenge. A significant decrease in hematocrit was evident post-challenge in unvaccinated fish only. (Bar = mean value/ Box = 95% confidence interval)

-78-

Figure 1b: Serum protein Serum protein values for vaccinated and unvaccinated fish, pre and post-challenge. A significant decrease in serum protein was evident after challenge in unvaccinated fish only. (Bar = mean value/ Box = 95% confidence interval)

-79-

Figure 2a: Leukocyte count variation with ECP exposure, pre and post-challenge. A significant increase in the total leukocyte count was evident post-challenge for fish exposed to the ECP, but not for unexposed fish. (Bar = mean value/ Box = 95% confidence interval)

-80-

Figure 2b: Leukocyte count variation with the chronology of the ECP-exposure. A significantly lower total leukocyte count was evident for fish exposed to the ECP 22 days post- vaccination compared tofish exposed 1 day prior to or simultaneously with vaccination. (Bar = mean value/ Box = 95% confidence interval)

-81-

Figure 3a: Small lymphocyte counts and ECP exposure. A significant increase in the small lymphocyte count was evident post-challenge for fish exposed to the ECP. (Bar = mean value/ Box = 95% confidence interval)

-82-

Figure 3b: Small lymphocyte count based upon the chronology of the ECP exposure. A significant decrease in small lymphocyte counts was evident in the fish exposed to the ECP 22 days after vaccination compared to fish exposed 1 day prior to or simultaneously with vaccination. (Bar = mean value/ Box = 95% confidence interval)

-83-

Figure 3c: Small lymphocyte count and vaccination status. A significant increase in small lymphocyte count was evident post-challenge for fish vaccinated against Y. ruckeri, but not for unvaccinated fish. (Bar = mean value/ Box = 95% confidence interval)

-84-

Figure 4: Large lymphocyte count.

A significant increase in the large lymphocyte differential cell count was evident post-challenge for unvaccinated fish, but not for vaccinated fish (Bar = mean value/ Box = 95% confidence interval)

-85-

Figure 5: PMN cell count. A significant decrease in the PMN differential cell count was evident post-challenge for vaccinated fish, while a significant post-challenge increase in this value was evident for the unvaccinated fish. (Bar = mean value/ Box = 95% confidence interval)

-86-

Figure 6a: Monocyte counts based upon ECP exposure. A significant increase in the differential monocyte count was evident post-challenge in fish not exposed to the ECP, but not for the ECP-exposed fish. (Bar = mean value/ Box = 95% confidence interval)

-87-

Figure 6b: Monocyte counts based upon vaccination status.

A significant increase in the monocyte differential count wasevident post-challenge only for unvaccinated fish.

(Bar = mean value/ Box = 95% confidence interval)

-88-

Figure 7: Microagglutination antibody titer. Statistical evaluation of the microagglutination titer showed a significant post-challenge increase in titer for vaccinated fish. (Bar = mean value/ Box = 95% confidence interval)

-89-

Figure 8: Phagocytosis.

Phagocytosis capabilites of splenic phagocytes of fish from all experimental groups. A statistically significant decrease in phagocytosis was evident post-challenge. (Bar = mean value/ Box = 95% confidence interval)

-90-

Figure 9: Survival. A survival estimate (proportion) for the two experimental groups with unvaccinated fish. ECP = ECP-injected, PBS = PBS-injected (control). Thiry days post-challenge, These two groups were the only to suffer ERM-related mortalities. While survival was higher for the ECP-injected group, the difference in survival between these two groups was not significant at p = 0.05. (Bar = mean value/ Box = 95% confidence interval)

-91-

References:

Alexander, J.B. and G.A. Ingram. 1992. Noncellularnonspecific defense mechanisms of fish. Annual Review of FishDiseases. 2:249-279.

Amend, D.F. and K.A. Johnson. 1984. Evidence for lack ofantigenic competition among various combinations of Vibrioanguillarum, Yersinia ruckeri, Aeromonas salmonicida, andRenibacterium salmoninarum bacterins when administered tosalmonid fishes. Journal of Fish Diseases. 7:293-299.

Anderson, D.P. 1992a. Chapter 8- Immunization of fishspleen sections and NBT, phagocytosis, PFC, and antibody assaysfor monitoring the immune response. In: J.S. Stolen, T.C.Fletcher, D.P. Anderson, S.L. Kaattari, and A.F. Rowley(Editors), Techniques in Fish Immunology, FITC-2. SOSPublications, Fair Haven, NJ, USA. pp. 79-87.

Anderson, D.P. 1992b. Immunostimulants, adjuvants, andvaccine carriers in fish: applications in aquaculture. AnnualReview of Fish Diseases. 2:249-280.

Anderson, D.P. and O.W. Dixon. 1989. Fish biologicsguide: regimens and protocols for the production and use ofantisera, antigens and other reagents for fish diseaseserodiagnostics. U.S. Fish and Wildlife Service, Kearneysville,WV. 215 pp.

Austin, B. and D.A. Austin. 1993. Bacterial FishPathogens, 2 ed. Ellis Horwood Limited, Inc., New York, NY. nd

384 p.

Bandin, I., A.E. Ellis, J.L. Barja, and C.J. Secombes. 1993. Interaction between rainbow trout macrophages andRenibacterium salmoninarum in vitro . Fish and ShellfishImmunology. 3:25-33.

Bandin, I., Y. Santos, D.W. Bruno, R.S. Raynard, A.E.Toranzo, and J.L. Barja. 1991. Lack of biological activitiesin the extracellular products of Renibacterium salmoninarum . Canadian Journal of Fisheries and Aquatic Sciences. 48:421-425.

Bandin, I., Y. Santos, B. Magarinos, J.L. Barja, and A.E.Toranzo. 1992. The detection of two antigenic groups amongRenibacterium salmoninarum isolates. FEMS Microbiology Letters. 94:105-110.

-92-

Bradford, M.M. 1976. A rapid and sensitive method forthe quantitation of protein utilizing the principle of protein-dye binding. Analytical biochemistry. 72:248-254.

Bullock, G.L., B.R. Griffin, and H.M. Stuckey. 1980. Detection of Corynebacterium salmoninus by direct fluorescentantibody test. Canadian Journal of Fisheries and AquaticSciences. 37:719-721.

Campbell, T.W. 1988. Fish cytology and hematology. TheVeterinary Clinics of North America, Small Animal Practice. 18(2):347-364.

Cipriano, R.C., L.A. Ford, J.D. Teska, and L.E. Hall. 1992. Detection of Aeromonas salmonicida in the mucus ofsalmonid fish. Journal of Aquatic Animal Health. 4:114-118.

Daly, J.G. and R.M.W. Stevenson. 1989. Agglutination ofsalmonid spermatozoa by Renibacterium salmoninarum . Journal ofAquatic Animal Health. 1:163-164.

Daly, J.G. and R.M.W. Stevenson. 1987. Hydrophobic andhemagglutinating properties of Renibacterium salmoninarum . Journal of General Microbiology. 133:3575-3580.

Dubreuil, J.D., M. Jacques, L. Graham, and R. Lallier. 1990. Purification and biochemical and structuralcharacterization of a fimbral hemagglutinin of Renibacteriumsalmoninarum. Journal of General Microbiology. 136:2443-2448.

Evendon, A.J., T.H. Grayson, M.L. Gilpin, and C.B. Munn. 1993. Renibacterium salmoninarum and bacterial kidney disease-the unfinished jigsaw. Annual Review of Fish Diseases. 3:87-104.

Fiedler, F. and R. Draxl. 1986. Biochemical andimmunochemical properties of the cell surface of Renibacteriumsalmoninarum . Journal of Bacteriology. pp. 799-804.

Getchell, R.G., J.S. Rohovec, and J.L. Fryer. 1985. Comparison of Renibacterium salmoninarum isolates by antigenicanalysis. Fish Pathology. 20(2/3):149-159.

Griffiths, S.G. and W.H. Lynch. 1991. Instability o themajor soluble antigen produced by Renibacterium salmoninarum . Journal of Fish Diseases. 14:55-66.

-93-

Hsu, H.M., P.R. Bowser, and J.H. Schachte, Jr. 1991. Development and evaluation of a monoclonal-antibody-based enzyme-linked immunosorbent assay for the diagnosis of Renibacteriumsalmoninarum infection. Journal of Aquatic Animal Health. 3:168-175.

Kaattari, S.L. 1992. Fish B lymphocytes: defining theirform and function. Annual Review of Fish Diseases. 2:161-180.

Kaattari, S.L., P. Turaga, and G. Wiens. 1989. Development of a vaccine for bacterial kidney disease in salmon -Final report. Bonneville Power Administration, Portland, OR. 310 p.

Klontz, G.W. 1994. Fish hematology. In: J.S. Stolen,T.C. Fletcher, A.F. Rowley, J.T. Zelikoff, S.L. Kaattari, andS.A. Smith (editors), Techniques in Fish Immunology - 3. SOSPublications, Fair Haven, NJ, USA. pp. 121-130.

Rockey, D.D., P.S.D. Turaga, G.D. Wiens, B.A. Cook, and S.L.Kaattari. 1991. Serine proteinase of Renibacteriumsalmoninarum digests a major autologous extracellular and cellsurface protein. Canadian Journal of Microbiology. 37:758-763.

Sami, S., T. Fischer-Scherl, R.W. Hoffman, and C. Pfeil-Putzien. 1992. Immune complex-mediated glomerulonephritisassociated with bacterial kidney disease in the rainbow trout( Oncorhynchus mykiss ) Veterinary Pathology. 29:169-174.

Schill, W.B., S.R. Phelps, and S.W. Pyle. 1984. Multilocus electrophoretic assessment of the genetic structureand diversity of Yersinia ruckeri . Applied and EnvironmentalMicrobiology. 48(5):975-979.

Secombes, C.J. and T.C. Fletcher. 1992. The role ofphagocytes in the protective mechanisms of fish. Annual Reviewof Fish Diseases. 2:53-71.

Shieh, H.S. 1988. An extracellular toxin produced byfish kidney disease bacterium, Renibacterium salmoninarum . Microbios. Letters. 38:27-30.

Tizard, I. 1987. Veterinary Immunology - 3 edition. rd

W.B. Saunders Co., Philadelphia, PA. 401 p. Turaga, P.S.D., G.D. Wiens, and S.L. Kaattari. 1987a.

Analysis of Renibacterium salmoninarum antigen production insitu . Fish Pathology. 22(4):209-214.

-94-

Turaga, P.S.D., G.D. Wiens, and S.L. Kaattari. 1987b. Bacterial kidney disease; the potential role of soluble proteinantigen(s). Fish Biology. 37(Supplement A):191-194.

Wiens, G.D. and S.L. Kaattari. 1989. Monoclonal antibodyanalysis of common surface protein(s) of Renibacteriumsalmoninarum . Fish Pathology. 24(1):1-7.

Wolf, K. and M.C. Quimby. 1973. Fish viruses: buffersand methods for plaquing eight agents under normal atmosphere. Applied Microbiology. 25:659-664.

Yasutake, W.T. and J.H. Wales. 1983. Microscopic anatomyof salmonids: an atlas. Resource Publication 150. U.S.Department of the Interior, Fish and Wildlife Service,Washington, D.C. 190 p.

Acknowledgments:I wish to thank my co-investigators in this investigation,

Dr. Larisa Ford, Ms. Cynthia Baker, and Dr. Stephen Smith. Wealso would like to thank the investigators and technical staff ofthe National Fish Health Research Laboratory, Kearneysville, WVfor their support with this investigation. We especially wish toacknowledge Rocco Cipriano, Cliff Starliper, and Phil McAllisterfor their contributions. We also thank David Smith of theAquatic Ecology Laboratory, Kearneysville, WV for his assistancewith the statistical analyses.

-95-

Appendices: Data Summary Tables

Data summaries for the immunological parameters evaluated arecontained in the following tables of this appendix. Numbers incells of the tables represent the mean values for all fish sampledat each sampling interval in each treatment group, followed by thestandard deviations (in parentheses). For the two groups in whichthe ECP was given 8 and 22 days post-vaccination, all pre-challengesample times are based upon ECP injection alone, and thereforethese pre-challenge samples are not chronologically equivalent tothe other groups.

Appendix A............HematocritAppendix B............Total Serum ProteinAppendix C............Microagglutination Titer *Appendix D............Total Leukocyte CountAppendix E............Small Lymphocyte CountAppendix F............Large Lymphocyte CountAppendix G............Polymorphonuclear Cell CountAppendix H............Monocyte CountAppendix I............Phagocytosis

( * Microagglutination titer numerical data is based on reciprocalsof log dilutions. Mean values with standard deviations are not2represented in Appendix C; instead, all titer values for eachexperimental group and sampling interval are listed.)

-96-

Appendix A: Hematocrit (%)

vaccinated unvaccinated ECP given at: time

ECP no ECP no -1 d 0 d 8 d 22 dECP ECP

12 h 3.8 34.6 39.4 40.7 34.2 33.8 37.3 30.9(3.1) (3.8) (4.8) (3.7) (5.8) (3.1) (8.8) (4.3)

1 d 35.3 32.4 39.6 38.8 35.8 35.3 37.8 33.3(3.7) (2.9) (3.7) (5.7) (3.9) (3.7) (4.4) (5.7)

2 d 39.3 38.4 40.3 42.4 34.9 39.3 43.2 35.6(7.1) (4.4) (4.5) (3.9) (4.9) (7.1) (4.2) (5.5)

4 d 36.1 38.4 36.7 41.3 36.1 36.1 39.6 34.8(3.5) (5.3) (4.2) (6.4) (4.5) 3.5) (6.1) (5.2)

8 d 39.0 38.8 37.4 36.1 39.3 39.0 35.4 33.6(4.1) (7.6) (4.2) (4.1) (5.6) (4.1) (6.4) (6.8)

16 d 34.8 36.7 36.6 40.1 36.3 34.8 35.6 34.4(3.2) (6.8) (2.1) (5.2) (7.1) (3.2) (2.9) (4.8)

22 d 34.7 35.8 37.6 39.3 38.3 34.7 --- ---(4.3) (3.5) (7.0) (5.7) (9.3) (4.3)

* Time of Challenge

23 d 35.7 35.1 33.3 35.3 33.6 35.7 30.8 33.3(4.3) (4.0) (3.6) (5.5) (2.0) (4.3) (6.0) (5.7)

26 d 37.0 39.3 36.7 37.1 34.0 37.0 36.8 34.8(4.9) (4.3) (4.8) (6.9) (3.3) (4.9) (6.4) (5.2)

30 d 40.3 40.6 30.6 33.1 38.1 40.3 39.4 33.6(7.3) (2.1) (4.4) (3.2) (5.5) (7.3) (5.5) (6.8)

38 d 31.1 35.1 30.0 25.3 39.7 31.1 36.0 34.4(2.1) (4.0) (5.6) (6.2) (6.9) (2.1) (2.7) (4.8)

52 d 34.8 32.7 28.4 34.0 36.6 34.8 39.4 37.1(3.6) (4.1) (8.4) (7.8) (6.8) (3.6) (3.1) (5.3)

-97-

Appendix B: Total serum protein (g/dl)

ECP given at: time

vaccinated unvaccinated

ECP -1 d 0 d 8 d 22 d no ECP no ECP ECP

12 h 3.7 3.8 4.3 4.6 3.4 3.7 3.1 4.2(0.8) (0.8) (0.7) (0.2) (0.9) (0.8) (1.1) (0.7)

1 d 4.1 3.3 4.7 4.3 3.8 4.1 3.6 3.6(0.7) (0.6) (0.1) (0.3) (0.8) (0.7) (0.6) (0.7)

2 d 4.1 3.5 4.3 4.2 3.3 4.1 3.4 3.9(1.1) (0.6) (0.4) (0.5) (0.5) (1.1) (0.5) (0.8)

4 d 3.4 4.0 3.7 3.9 3.6 3.4 3.4 3.5(0.4) (0.6) (0.3) (0.4) (0.4) (0.4) (0.8) (0.9)

8 d 3.8 4.1 3.7 3.8 3.4 3.8 3.8 4.1(0.6) (1.2) (0.6) (0.9) (0.6) (0.6) (0.4) (0.4)

16 d 3.3 3.5 3.6 3.7 3.5 3.3 3.6 4.0(0.7) (1.0) (0.5) (0.8) (0.7) (0.7) (0.8) (0.8)

22 d 3.8 3.7 3.7 4.0 3.9 3.8 --- ---(0.4) (0.5) (0.6) (0.6) (0.9) (0.4)

* Time of challenge

23 d 4.0 4.1 3.8 3.3 3.8 4.0 3.3 3.6(0.5) (0.6) (0.6) (0.7) (0.4) (0.5) (0.5) (0.7)

26 d 3.8 3.5 3.5 3.5 2.9 3.8 3.6 3.5(0.4) (0.7) (0.3) (0.4) (0.9) (0.4) (0.5) (0.9)

30 d 4.4 3.8 3.3 3.4 3.4 4.4 4.1 4.1(0.5) (0.2) (0.4) (0.6) (0.7) (0.5) (0.4) (0.4)

38 d 4.3 4.7 4.0 4.3 4.4 4.3 4.2 4.0(0.2) (0.6) (0.8) (0.5) (0.6) (0.2) (0.3) (0.8)

52 d 3.6 3.4 3.1 3.9 3.4 3.6 3.9 4.2(0.3) (0.5) (0.7) (0.4) (0.9) (0.3) (0.3) (0.3)

-98-

Appendix C: Microagglutination antibody titers

Time Vaccinated ECP given at:Unvaccinated

ECP no ECP no 0 d 8 d 22 dECP ECP

-1 d

12 h 2,1,1,1, 1,1,3,0, 2,1,0,4, 0,0,0,0, 2,2,1,2, 2,1,1,1, 2,0,0,0, 1,0,2,7 1,1,1 3,1,2 1,4,1 0,1,1 1,0,3 1,1,1 0,1,0 6,2,5

1 d 1,0,1,0, 0,0,0,1, 0,0,0,1, 0,0,0,2, 0,0,0,2, 1,0,1,0, 0,0,2,0, 0,0,2,4, 0,1,0 0,0,1 0,1,3 0,0,2 3,2,2 0,1,0 0,2,0 4,0,1

2 d 0,0,0,1, 0,0,0,0, 0,1,0,0, 0,0,0,0, 0,2,0, 0,0,0,1, 0,0,0,0, 0,5,7,5, 1,0,0 2,2,1 1,0,1 0,1,1 0,0, 1,0,0 0,0,0 5,0,3

4 d 1,0,0,1, 1,0,0, 0,0,0,0, 0,0,0, 0,0,0,0, 1,0,0,1, 1,0,0, 0,0,0,0, 1,0,1 0,1,0 0,2,0 0,1,0 0,0,0 1,0,1 0,1,0 0,0,0

8 d 1,0,0,0, 0,0,1, 0,0,0,0, 0,1,0,0, 0,0,0, 1,0,0,0, 0,0,0,0, 5,1,0,6, 1,0,0 0,0,0 0,0,0 0,0,0 0,0,0 1,0,0 0,0,0 3,2,0

16 d 0,0,3,1, 2,0,1,4, 0,0,0, 0,0,0, 0,3,0, 0,0,3,1, 0,1,0, 5,5,3,3, 3,3,2 0,0,1 0,0,1 0,0,1 1,3 3,3,2 4,3 7,4,6

22 d 2,0,0,5, 1,4,1,0, 0,1,0,0, 0,1,0,1, 0,0,4,0, 2,0,0,5, --- --- 0,2,5 0,0,1 1,0,0 1,2,0 3,0,4 0,2,5

*Time of Challenge

23 d 1,1,1,0, 1,6,6,0 0,0,0,0, 0,0,0,0 0,1,1,0, 1,1,1,0, 1,1,6,3, 0,0,2,4, 4,0,1 7,0,4 0,0,0 0,1,0 0,2,7 4,0,1 3,6,2 4,0,1

26 d 6,5,7,0, 1,2,1,3, 0,0,1,0, 0,0,1,2, 4,6,3,1, 6,5,7,0, 6,6,6,2, 4,1,5,4, 6,4,2 3,6,7 0,0,2 0,0,0 4,3,4 6,4,2 3,2,5 0,4,5

30 d 0,5,2,6, 1,1,5,0, 7,1,4,1, 0,1,0,0, 5,4,3,4, 0,5,2,6, 5,1,1,6, 5,1,0,6, 5,4,6 5,2,5 1,0,1 1,1,1 5,6,4 5,4,6 1,0,4 3,2,0

38 d 1,7,7,1, 6,0,4,2, 0,3,0,0, 0,0,0,0, 0,0,5,5, 1,7,7,1, 7,0,5,4, 5,5,3,3, 4,5,5 3,1,1 0,0,2 0,0,0 7,5,6 4,5,5 3,8,5 7,4,6

52 d 0,0,4,4, 3,5,3,4, 3,1,2,3, 0,0,1,0, 4,0,0,3, 0,0,4,4, 2,0,4,3, 1,1,0,0, 5,6,6 1,5,10 2,1,2 0,3,2 5,2,5 5,6,6 4,1,7 3,4,3

-99-

Appendix D: Total Leukocyte Count (million cells/ml)

Time Unvaccinated Vaccinated ECP given at:

ECP no -1 d 22 dECP

ECP no 0 d 8 dECP

12 h 9.7 11.3 18.3 17.1 12.8 9.8 18.4 18.0(2.8) (5.3) (6.3) (5.2) (3.5) (2.8) (5.7) (5.9)

1 d 12.6 14.7 13.3 9.5 16.1 12.6 35.3 17.5(7.9) (6.9) (6.9) (6.2) (5.4) (7.9) (8.5) (6.1)

2 d 17.2 19.3 18.2 18.1 30.1 17.2 37.2 17.3(5.3) (7.5) (6.6) (5.8) (7.9) (5.3) (14.0 (10.2)

4 d 21.5 26.7 21.3 23.3 13.1 21.5 34.4 27.2(3.7) (10.1) (11.3) (8.1) (6.5) (3.7) (12.9 (4.3)

8 d 23.1 20.1 16.8 27.6 19.0 23.1 22.2 21.8(10.5) (8.5) (8.0) (10.1) (9.9) (10.5) (3.1) (5.8)

16 d 27.6 27.7 20.0 26.6 27.7 27.6 20.3 19.3(16.1) (14.7) (7.7) (10.0) (9.1) (16.1) (6.8) (5.7)

22 d 25.2 24.3 27.8 18.7 33.1 25.2 --- ---(9.9) (11.0) (5.3) (6.1) (13.0) (9.9)

* Time of Challenge

23 d 24.8 15.8 28.6 23.3 43.6 24.8 17.7 17.5(11.7) (4.6) (12.6) (9.5) (13.9) (11.7) (5.6) (6.1)

26 d 38.8 15.8 28.6 23.3 43.6 24.8 17.7 17.5(19.1) (4.6) (12.6) (9.5) (13.9) (11.7) (5.6) (6.1)

30 d 34.1 30.3 17.7 14.7 31.5 34.1 34.9 21.8(18.6) (6.2) (6.1) (5.0) (10.6) (18.6) (11.3) (5.8)

38 d 16.8 10.0 13.7 24.2 23.2 16.8 12.8 19.3(8.4) (9.5) (6.7) (9.8) (12.0) (8.4) (2.1) (5.7)

52 d 44.7 35.9 33.9 28.3 18.8 44.7 32.5 29.1(12.8) (16.9) (11.2) (15.3) (7.3) (12.8) (10.5) (7.7)

-100-

Appendix E: Small Lymphocyte Count (million cells/ml)

Time Unvaccinated Vaccinated ECP given at:

ECP ECP -1 d 8 d 22 d no no 0 dECP ECP

12 h 8.3 9.5 15.1 16.2 8.7 8.3 15.4 14.0(2.7) (4.8) (5.2) (5.3) (4.1) (2.7) (5.8) (5.2)

1 d 9.1 9.6 10.8 9.0 11.7 9.1 29.0 14.9(7.0) (5.2) (5.4) (5.9) (6.0) (6.9) (8.4) (5.4)

2 d 12.0 13.0 17.2 16.5 24.1 12.0 31.8 15.5(4.6) (5.7) (6.2) (5.9) (5.6) (4.6) (12.3) (9.3)

4 d 18.4 22.8 19.7 22.1 10.7 18.4 31.4 25.4(3.3) (9.0) (11.2) (8.0) (4.9) (3.3) (11.6) (3.6)

8 d 24.4 18.9 15.6 26.3 16.7 24.4 21.4 21.0(7.2) (8.4) (7.5) (9.7) (9.2) (9.2) (3.5) (6.2)

16 d 26.9 26.2 18.8 25.5 26.4 26.9 17.8 18.3(16.2) (14.0) (8.0) (9.7) (8.6) (16.2) (5.7) (5.3)

22 d 21.4 22.6 25.8 17.1 30.7 21.4 --- ---(7.2) (10.3) (5.8) (5.8) (13.1) (7.2)

* Time of Challenge

23 d 23.1 14.5 25.2 19.5 40.6 23.1 15.5 14.9(10.7) (4.1) (10.0) (10.2) (13.7) (10.7) (5.3) (5.4)

26 d 37.7 22.4 40.0 25.2 31.6 37.8 35.3 25.4(18.6) (10.7) (10.8) (5.4) (7.5) (18.6) (14.1) (3.6)

30 d 33.0 28.6 15.5 11.7 30.2 33.0 33.0 21.0(19.1) (6.4) (5.8) (5.0) (10.4) (19.1) (6.0) (6.2)

38 d 15.3 9.3 10.7 18.1 21.8 15.3 11.9 18.3(7.8) (8.6) (7.0) (10.6) (11.0) (7.8) (2.4) (5.3)

52 d 42.9 34.8 28.1 28.1 18.1 42.9 31.7 27.7(12.1) (16.5) (7.9) (13.6) (6.9) (12.1) (10.2) (7.9)

-101-

Appendix F: Large Lymphocyte Count (million cells/ml)

Time Unvaccinated ECP given at: Vaccinated


12 h 0.12 0.18 0.42 0.29 0.88 0.12 0.55 1.22(0.08) (0.28) (0.22) (0.18) (0.93) (0.08) (0.73) (0.59)

1 d 0.71 0.65 0.59 0.18 0.54 0.71 0.91 0.86(0.61) (0.59) (0.58) (0.14) (0.39) (1.04)(0.61) (0.70)

2 d 0.34 0.66 0.12 0.35 0.70 0.34 0.54 0.39(0.23) (0.12) (0.21) (0.34) (0.25)(0.84) (0.23) (0.68)

4 d 0.37 0.69 0.14 0.40 0.56 0.37 0.38 0.47(0.29) (0.88) (0.22) (0.56) (0.29) (0.75) (0.45)(0.18)

8 d 0.23 0.44 0.42 0.10 0.62 0.23 0.31 0.35(0.26) (0.31) (0.14) (0.54) (0.26) (0.43) (0.22)(0.40)

16 d 0.12 0.24 0.29 0.51 0.60 0.12 0.41 0.25(0.13) (0.25) (0.19) (0.59) (0.63) (0.13) (0.60) (0.23)

22 d 0.18 0.55 0.72 0.44 0.39 0.18 --- ---(0.21) (0.54) (0.58) (0.59) (0.47) (0.21)

* Time of Challenge

23 d 0.20 0.46 0.46 0.69 1.29 0.20 0.46 0.86(0.26) (0.54) (0.40) (0.22) (0.63) (0.26) (0.32) (0.70)

26 d 0.07 0.14 0.15 0.55 0.86 0.07 0.70 0.47(0.14) (0.25) (0.25) (0.50) (0.46) (0.14) (0.77) (0.45)

30 d 0.30 0.46 1.05 1.16 0.25 0.30 0.50 0.35(0.34) (0.38) (1.52) (0.92) (0.29) (0.34) (0.62) (0.22)

38 d 0.26 0.19 1.19 2.09 0.11 0.26 0.18 0.25(0.39) (0.33) (1.86) (0.19) (0.39) (0.15) (0.23)(1.80)

52 d 0.36 0.12 1.62 0.80 0.27 0.36 0.11 0.21(0.40) (0.23) (3.07) (0.77) (0.23) (0.40) (0.19) (0.27)

-102-

Appendix G: Polymorphonuclear Cell Count (million cells/ml)



12 h 1.27 1.61 2.82 0.70 2.50 1.27 2.26 2.68(0.46) (1.48) (2.92) (0.41) (2.64) (0.46) (1.91) (1.33)

1 d 2.63 4.18 1.84 0.37 3.72 2.63 4.95 1.43(2.00) (3.65) (1.64) (0.29) (1.24) (2.00) (3.22) (0.55)

2 d 4.69 5.24 0.79 1.23 5.07 4.69 4.68 1.43(3.33) (4.07) (0.77) (0.93) (3.23) (3.33) (2.49) (1.68)

4 d 2.49 3.05 1.45 0.71 1.77 2.49 2.68 1.28(1.77) (1.93) (1.73) (0.39) (1.62) (1.77) (3.76) (1.14)

8 d 0.85 0.65 0.74 1.17 1.34 0.85 0.52 0.86(0.79) (0.50) (0.36) (1.10) (1.68) (0.79) (0.40) (0.76)

16 d 0.56 1.15 0.89 0.58 0.68 0.56 1.07 0.73(0.86) (0.77) (0.96) (0.41) (0.56) (0.86) (1.57) (0.53)

22 d 0.80 1.13 1.26 1.17 1.90 0.80 --- ---(0.83) (0.72) (0.64) (1.26) (0.90) (0.83)

* Time of Challenge

23 d 1.52 0.72 2.73 2.24 1.33 1.52 1.52 1.43(1.82) (0.35) (3.11) (1.28) (0.90) (1.82) (1.09) (0.55)

26 d 1.01 0.31 1.18 0.99 2.24 1.01 1.13 1.28(1.52) (0.25) (0.60) (0.23) (2.47) (1.52) (1.25) (1.14)

30 d 0.81 1.25 1.22 1.74 0.99 0.81 1.40 0.86(1.18) (1.17) (0.93) (1.28) (0.99) (1.18) (0.78) (0.76)

38 d 1.26 0.49 1.64 3.41 1.19 1.26 0.69 0.73(0.90) (0.62) (1.12) (2.56) (1.04) (0.90) (0.69) (0.53)

52 d 1.35 0.85 3.99 1.26 0.45 1.35 0.71 1.09(0.98) (0.52) (3.22) (1.50) (0.54) (0.98) (0.69) (0.69)

-103-

Appendix H: Monocyte Count (million cells/ml)


ECP no ECP no -1 d 0 d 8 d 22 d ECP ECP

12 h 0.02 0 0.03 0.03 0.12 0.02 0.16 0.02(0.05) (0) (0.07) (0.07) (0.27) (0.05) (0.27) (0.05)

1 d 0.21 0.29 0.09 0 0.09 0.21 0.48 0.26(0.30) (0.40) (0.18) (0) (0.15) (0.30) (0.59) (0.19)

2 d 0.18 0.34 0.04 0.04 0.20 0.18 0.15 0.04(0.31) (0.41) (0.10) (0.10) (0.32) (0.31) (0.26) (0.11)

4 d 0.26 0.13 0 0.06 0.14 0.26 0 0.08(0.50) (0.36) (0) (0.10) (0.31) (0.50) (0) (0.13)

8 d 0 0.07 0.05 0 0.03 0 0 0 (0) (0.13) (0.10) (0) (0.05) (0) (0) (0)

16 d 0 0.08 0.05 0 0 0 0 0.06 (0) (0.20) (0.10) (0) (0) (0) (0) (0.11)

22 d 0 0.03 0.08 0 0.08 0 --- --- (0) (0.08) (0.14) (0) (0.14) (0)

* Time of Challenge

23 d 0.02 0.09 0.22 0.52 0.36 0.02 0.22 0.26(0.04) (0.11) (0.34) (0.25) (0.96) (0.04) (0.28) (0.19)

26 d 0 0.28 0.05 0.14 0.18 0 0 0.08 (0) (0.32) (0.15) (0.19) (0.30) (0) (0) (0.13)

30 d 0 0 0.02 0.13 0.05 0 0 0 (0) (0) (0.04) (0.23) (0.12) (0) (0) (0)

38 d 0.05 0.02 0.17 0.56 0.10 0.05 0.04 0.06(0.09) (0.04) (0.26) (0.56) (0.17) (0.09) (0.06) (0.11)

52 d 0.08 0.07 0.16 0 0.06 0.08 0.05 0.10(0.22) (0.18) (0.33) (0) (0.12) (0.22) (0.13) (0.26)

-104-

Appendix I: Phagocytosis (%)

Time Vaccinated Unvaccinated ECP given at:

ECP no ECP no -1 d 0 d 8 d 22 d ECP ECP

12 h 8.6 12.4 16.8 16.8 18.4 8.6 6.8 3.8(5.4) (5.4) (5.8) (8.8) (5.9) (5.4) (2.2) (1.6)

1 d 11.0 7.8 10.6 11.4 11.2 11.0 5.4 7.4(5.1) (5.8) (6.0) (1.3) (5.9) (5.1) (3.0) (4.3)

2 d 11.6 6.6 14.0 8.6 4.8 11.6 7.2 5.6(1.7) (2.9) (4.2) (5.3) (4.0) (1.7) (3.0) (2.3)

4 d 5.6 7.6 6.6 10.0 14.4 5.6 5.2 7.8(3.1) (4.9) (3.1) (5.6) (3.4) (3.1) (3.4) (5.4)

8 d 1.2 10.6 8.6 5.4 11.6 1.2 7.4 2.0(1.8) (6.3) (4.9) (4.0) (9.8) (1.8) (4.3) (1.2)

16 d 6.0 10.2 5.2 6.4 13.6 6.0 5.2 10.2(2.1) (7.6) (4.2) (3.4) (6.3) (2.1) (6.6) (6.2)

22 d 9.4 17.8 4.4 5.2 10.0 9.4 --- ---(5.3) (11.3 (3.4) (3.3) (5.3)(3.2)

* Time of Challenge

23 d 9.6 11.0 8.8 5.2 4.4 9.6 11.0 7.4(8.1) (4.1) (5.8) (4.3) (4.5) (8.1) (7.4) (4.3)

26 d 11.0 10.4 3.4 3.6 9.4 11.0 12.4 7.8(5.5) (8.9) (4.9) (4.8) (6.1) (5.5) (8.8) (5.9)

30 d 3.8 2.2 9.4 14.8 4.8 3.8 3.0 2.0(2.7) (1.6) (9.5) (9.8) (3.8) (2.7) (2.9) (1.2)

38 d 3.2 11.2 8.8 8.0 4.8 3.2 5.2 10.2(1.9) (1.9) (3.9) (5.6) (4.1) (1.9) (1.3) (6.2)

52 d 3.6 1.2 0 2.0 0.4 3.6 1.8 1.2(1.1) (1.3) (0) (0.9) (1.1) (2.0) (1.6)(1.9)

-105-

Chapter 3

In vitro effects of the extracellular protein of Renibacterium salmoninarum on phagocyte function

in brook trout ( Salvelinus fontinalis )

-106-

Abstract

Renibacterium salmoninarum is a facultative intracellular pathogenoften found in host phagocytes where it appears to successfullyavoid the host fish's immunological defenses. The objective ofthis investigation was to determine whether the solubleextracellular protein produced by R. salmoninarum may contribute tothe immunomodulation in bacterial kidney disease via inhibition ofhost phagocyte respiratory burst and/or phagocytosis mechanisms.Splenic cells from healthy adult brook trout ( Salvelinusfontinalis ) were incubated with two different concentrations ofextracellular protein or viable R. salmoninarum . Splenic cellcultures were then evaluated for respiratory burst activity viaflow cytometry with the dichlorofluorescin diacetate assay and forphagocytosis via light microscopic assessment of microsphereengulfment. Respiratory burst activity was inhibited in allexperimental groups as compared to controls, while no differenceswere noted in phagocytic abilities in cells among the treatmentgroups.

-107-

I. IntroductionRenibacterium salmoninarum , a Gram-positive diplobacillus, is

the causative agent of bacterial kidney disease (BKD) of salmonidfish. Affecting various species of both feral and culturedsalmonids, BKD is a chronic granulomatous disease producing highlevels of morbidity and mortality.

Bacterial kidney disease pathogenesis is complex, involvingimmunomodulatory effects of the pathogen upon the host.Renibacterium salmoninarum , similar to other pathogens such asMycobacterium spp. and Listeria spp., has the ability to exist andeven multiply intracellularly inside macrophages. Young andChapman (1978) first described the intracellular existence of thisbacteria, prior to its classification as Renibacterium . Flano etal. (1996) further characterized the chronology of BKDimmunopathogenesis, including the intracellular existence of R.salmoninarum in hematopoietic tissue of experimentally infectedcoho salmon ( Oncorhynchus kisutch ). The ability of R. salmoninarumto resist killing and thereby survive and even replicateintracellularly inside rainbow trout macrophages over the course ofseveral days was also observed by Klein (1992) and Bandin et al.(1993a). The significance of these findings to the pathogenesis ofBKD is largely undetermined. Evendon et al. (1993) suggested thatthe enhanced intracellular survival of R. salmoninarum inphagocytic cells supports the likelihood that cellular transportplays a role in the pathogen's dissemination. However, Flano etal. (1996) proposed that the bacteria merely incubatesintracellularly for a period whereas systemic dissemination occursvia freely circulating organisms.

In recent years, many investigations have examined theimmunomodulatory effects of R. salmoninarum on salmonid phagocytes.Rose and Levine (1992) found that complement-mediated opsonizationof R.salmoninarum enhanced bacterial adherence to and phagocytosisby rainbow trout macrophages. Similarly, Bandin et al. (1993b)found that bacteria coated with serum had enhanced survival insiderainbow trout phagocytes. Brattgjerd et al. (1996) determined thatthe immunostimulatory effects of M-Glucan on salmonid macrophageactivities including phagocytosis did not extend to enhancement ofmacrophage uptake of R. salmoninarum in Atlantic salmon ( Salmosalar ). Investigations describing the respiratory burst activity ofmacrophages with regard to R. salmoninarum revealed that thesuperoxide anion response of rainbow trout macrophages wassignificantly decreased in the presence of R. salmoninarum in vitro(Bandin et al., 1993a). However, R. salmoninarum organisms aresusceptible to killing by free radicals (H O ) in vitro (Hardie et2 2al., 1994). Sakai et al. (1996) found a significantly lowerchemiluminescent response of salmonid phagocytes against R.salmoninarum when compared against a non-salmonid fish, indicatingpotentially diminished ability of salmonid phagocyte-mediatedkilling of R. salmoninarum .

-108-

The extracellular protein (ECP) produced by R. salmoninarumhas also received considerable attention in the literature withregard to its role in BKD pathogenesis. The major soluble antigenof the ECP is most widely reported to be a relatively unstable 57kD antigen found as a cell surface component or as a free protein,although discrepancies exist in the literature regarding stabilityand molecular weight (Griffiths and Lynch, 1991; Fielder and Draxl,1986). Renibacterium salmoninarum ECP has been predictablyencountered in association with both artificial microbial culturesand BKD-infected salmonid tissues. Tissue concentrations of ECP ashigh as 1.0 mg/ml have been reported (Kaattari et al., 1989). Bothin vivo and in vitro , the ECP has been associated with a number ofpotential immunomodulatory effects. Among these arehemagglutination of rabbit erythrocytes (Daly and Stevenson, 1987),decreased in vitro antibody response of coho salmon, decreasedhematocrit (Turaga et al., 1987), leukagglutination of salmonidleukocytes, and activation of macrophages (Kaattari et al., 1989).Still others have reported a complete lack of biological activityof the ECP (Bandin et al., 1991).

The purpose of this investigation was to determine whether theECP of Renibacterium salmoninarum directly affects the function ofsalmonid phagocytes in vitro . In particular, respiratory burstactivity and phagocytosis, two critical physiological processes ofcell-mediated immunity, were examined in splenic phagocytes exposedto R. salmoninarum extracellular protein.

II. Materials and Methods

A. Extraction of Renibacterium salmoninarum extracellular proteinExtracellular protein in solution was recovered by a

modification of the methods of Hsu et al., 1991 and Rockey et al.,1991. An isolate of the ATCC #33209 strain of R. salmoninarumobtained from the National Fish Health Research Laboratory inKearneysville, West Virginia was subcultured and incubated on KDM2agar plates at 15 C for 20 days. Samples from the cultures wereo

Gram-stained and serologically tested via the direct fluorescentantibody test (DFAT) for R. salmoninarum (Bullock et al., 1980) inorder to confirm pure cultures of R. salmoninarum . Ten plates wererinsed with approximately 24 ml of sterile water, and the bacterialcolonies were agitated to remove them from the agar. The bacterialsuspension was centrifuged for 20 min. at 3000 xg and 10 C.o

Twenty-two milliliters of supernatant were recovered and filtersterilized through a 0.2 µm filter (Acrodisc, Gelman Sciences;AnnArbor, MI). The bacterial cellular pellet was weighed todetermine the protein:cell ratio. The supernatant fractioncontaining the ECP was frozen at - 20 C for three weeks. Upono

initiation of the experiment, the solution was thawed and

-109-

maintained at 4 C. Protein concentration was determined with ao

modified enhanced BCA assay (Pierce; Rockford, IL), and the pH ofthe solution was determined with a standard pH meter.

B. Collection and incubation of splenic phagocytesSixteen healthy adult brook trout, Salvelinus fontinalis , from

stock maintained at the National Fish Health Research Laboratoryweighing approximately 200 g each (approximate splenic weights=0.5g) were sacrificed via a lethal dose of tricaine methanesulfonate(MS-222; Sigma Chemical Co., St. Louis, MO). The spleens wereaseptically removed and placed in cold (4 C) sterile Eagle'so

Minimum Essential Media (EMEM; Earle's salts containing 2% newborncalf serum, 100 mg/ml gentomycin, and 0.016 M Tris; Wolf andQuimby, 1973). Splenic cell suspensions were prepared via amodification of the method of Anderson (1992). Briefly, spleenswere finely macerated and the cellular debris was resuspended incold, sterile EMEM. Cell suspensions were washed twice in EMEM viacentrifugation for 5 min. at 100 xg and 4 C. Following the secondo

wash, cells were resuspended in 1.01 ml of EMEM. Ten microliters ofthe suspensions were removed for determination of approximateleukocyte concentration with a hemacytometer (Klontz, 1994).Concentrations of all 16 trout cell cultures were adjusted to 10million leukocytes/ml with additional EMEM and a minimum of 2 mltotal volume per sample. Each cell culture was then divided into four aliquots of 0.5 ml, arbitrarily labeled Groups A through D. Treatments (0.5 ml each) were added to each group as follows:

Group A: sterile phosphate buffered saline

Group B: ECP solution at a final (incubation) concentration of 0.1 mg/ml

Group C: ECP solution at a final concentration of 1.0 mg/ml

Group D: R. salmoninarum cells in phosphate buffered saline,equivalent to the production of 0.1 mg of ECP.

Treated cell cultures were incubated with gentle agitation inpolysterene round-bottomed tubes for approximately 18 h. at 4 C. o

C. Determination of phagocyte H O production in respiratory burst2 2activity

Respiratory burst assessment was performed via the method ofBass et al. (1983) as modified by Holladay et al. (1996).Following incubation, ten of the splenic cell cultures in each ofGroups A through D were prepared for evaluation of phagocyterespiratory burst activity with the dichlorofluoroescin diacetate(DCF-DA) assay for flow cytometry. A 200 µl aliquot from each ofthe 40 samples was incubated in a polysterene round-bottomed tube

-110-

for 15 min. at 4 C with 5 µl of DCF-DA (Molecular Probes, Eugene,o

OR; 5mM). Cells were subsequently stimulated by the addition of 10µl phorbol-12-myristate 13-acetate (PMA; Sigma, St. Louis, MO; 100ng/ml) and incubated similarly for 30 minutes. Immediatelyfollowing incubation, the splenic cell suspensions were filteredthrough a 45 µm nytex filter to remove large acellular debris priorto flow cytometry. Background fluorescence, determined usingunstained cells from each treatment group, was minimal and wassubtracted from respective populations incubated with thefluorescent probe.

Flow cytometric analysis was performed on a Coulter EPICS VFlow Cytometer interfaced with an MDADS data analysis computer(Coulter Electronics, Hialeah, FL) and a hard copy display providedby Tektronix 4612 Video Hard Copy Unit (Tektronix, Wilsonville,OR). Standardization of the flow cytometer was performed usingfluorescent calibration beads (Coulter) with 480 nm laser(Coherent, Palo Alto, CA) and excitation set at 300 mW. Apopulation of large cells containing phagocytes was isolated on atwo-parameter histogram of forward angle light scatter(FALS) andside angle light scatter(SALS), based on the high FALS (cell size)and SALS (granularity) characteristics of these cells. A 452-515laser blocker was employed to block laser light from otherfluorescent measurements. The fluorescent distribution wasdisplayed as a 256 channel, single parameter histogram. Greenfluorescence was measured through 550 nm longpass dichromic and 525band pass filters. A gated amplifier was used to electronicallyexclude small particles and cellular debris from evaluation. Foreach of the 40 treated samples and 4 unstained controls, 5,000events were collected and analyzed.

D. Determination of cellular phagocytosis The remaining six samples of Groups A through D were utilized

to evaluate cellular phagocytosis with a modification of themethods of Sakai et al. (1989) and Anderson (1992). Briefly,Fluoresbrite microspheres (1.16 µm; Polysciences, Inc., Warrington,PA) were adjusted to a concentration of 75,000 microspheres/µl insterile phosphate buffered saline, and 1 ml of the microspheresuspension was sonicated for 30 sec. to disrupt aggregation. Onemilliliter of additional EMEM was added to each of the 24 treatedsplenic cell cultures, followed by 20 µl of the microspheresuspension. Cells were resuspended with agitation and incubatedfor an additional 24 h. at 4 C. Following incubation, the splenico

cells were centrifuged for 5 min. at 100 xg and 10 C, washed ino

1.5 ml EMEM, and centrifuged again as before. The supernatant wasdiscarded, and the cell pellet was resuspended in 0.5 ml of EMEM.Each splenic cell culture was distributed evenly over two sterilemicroscope slides (0.25 ml/slide) and allowed to air dry at roomtemperature in the dark for 12 hours. Once dry, the slides were

-111-

fixed and stained using Leukostat (a Wright's stain modification;Fisher Diagnostics, Pittsburgh, PA).

Evaluation of phagocytosis was accomplished using conventionallight microscopy. Microscope slides were viewed with oil immersion(magnification = 100x), and phagocytic cells (monocytes/macrophagesand granulocytes) were identified based on morphology, anddifferentiated based on adherence/engulfment of microspheres(Figure 1). One hundred phagocytes were counted for each of the 24groups, and the percentage of cells phagocytosing microspheres wasrecorded for each sample. The count was repeated in triplicate, andthe mean percentage of phagocytosing cells was recorded for each ofthe samples.

III. Results

A. Extracellular protein Protein concentration in the twenty-two ml of the solution

containing the ECP was 2.0 mg/ml, yielding a total recovery of 44.0mg protein. The weight of the corresponding bacterial cell pelletwas 1.46 g, indicating a protein:cell ratio of 30.14 mg/g for thisculture. The pH of the protein solution was determined to be 8.05.

B. DCF-DA assay for Respiratory Burst ActivityIn the DCF-DA procedure, endogenous cellular esterases

hydrolyze acetate groups from the DCF-DA molecule following passivediffusion of parent compounds into cells. This process convertsDCF-DA into a non-fluorescent 2',7'-dichlorofluoroscene (DCFH)probe that is effectively trapped inside the cell. Subsequentproduction of H O by PMA-activated cells during the oxidative2 2respiratory burst oxidizes DCFH to the highly fluorescent 2',7'-dichlorofluoroscein (DCF) product, which can be readily detectedand quantitated by flow cytometry (Bass et. al., 1983).

Flow cytometric analysis of the ten samples from each of fourtreatment groups produced numerical values for mean levels ofcellular fluorescence as well as histograms displaying both meancellular fluorescence and dot-plot representations of forward andside angle light scatter for the gated cell population examined foreach sample. Statistical analysis of results revealed significantdifferences in the mean level of fluorescence (indicatingsignificant differences in respiratory burst activity ofphagocytes) in control Group A versus the three treatment groups(Dunnett's t-test, p< 0.05). Although there was no statisticallysignificant difference among the three treatment groups, meancellular fluorescence for Group C (treated with 1.0 mg/mlextracellular protein) was less than the fluorescence for Group B(0.1 mg/ml ECP) and Group D ( R. salmoninarum cells equivalent to

-112-

the cell weight producing 0.1 mg/ml ECP in the initial proteinharvest procedure). Groups B and D produced similar resultsnumerically and graphically (Table 1; Figure 2).

C. Phagocytosis of Fluorescent MicrospheresEvaluation of cellular phagocytosis for the six samples in

each of the four treatment groups revealed no significantdifferences in phagocytic capabilities among the four groups(Dunnett's t-test, p<0.05; Table 2; Figure 3).

IV. DiscussionAn understanding of the immunomodulatory effects of a pathogen

upon its host is critical to the development of protective measuresagainst disease. The majority of literature regarding immunologicalprophylactic control of BKD have emphasized humoral factors. Inrecent years, however, this emphasis is shifting as the importanceof cell-mediated protection in BKD pathogenesis is becomingapparent. For instance, Kaattari et al. (1989) reported that thegreatest resistance to BKD development during vaccine trialsoccurred in salmonids vaccinated with Freund's complete adjuvant,a non-specific macrophage activator. This finding, coupled with theconsistent difficulties encountered in attempts to develop avaccine for BKD utilizing humoral protective mechanisms (Evendon etal., 1993), suggests that additional focus on the cell-mediatedimmune response to R. salmoninarum is warranted.

Our investigation provides further evidence that theimmunomodulatory effects of R. salmoninarum may be exerted, atleast in part, through cell-mediated immunity. Specifically, theextracellular protein of the bacteria may inhibit phagocytic cellfunctions essential to their killing ability. Respiratory burstactivity, the production of oxygen free radicals in aerobicmicrobicidal defense, is a well established occurrence in piscinephagocytes, both mononuclear cells and granulocytes (Secombes andFletcher, 1992). Bandin et al. (1993a) observed that theintracellular survival of microbial pathogens is a virulencemechanism common to virulent and avirulent strains of R.salmoninarum , and that the macrophage superoxide anion response asstimulated by PMA was reduced in the presence of R. salmoninarum .These previous findings coupled with results of the presentinvestigation suggest that the inhibition of respiratory burstactivity via the extracellular protein may be related to R.salmoninarum 's resistance to intracellular killing by phagocytes.

This study provides no evidence of R. salmoninarum ECPinterference with phagocytosis. For a facultative intracellularpathogen which finds its way inside the host cell, interferencewith the phagocytosis processes of the cell may be

-113-

counterproductive to achieving intracellular existence. A number offacultative intracellular pathogenic bacteria, includingMycobacterium tuberculosis , Brucella abortus , and Listeriamonocytogenes are readily phagocytosed by host cells and rely onchemical interference or protective cell wall components to avoidintracellular killing. Moreover, certain properties of R.salmoninarum have been reported as potential enhancement factors inphagocytosis, including the bacteria's hydrophobicity,leukagglutination, and complement-mediated opsonization (Rose andLevine, 1992).

An interesting feature of our current results is thesuggestion of a dose-dependent relationship between theconcentration of ECP and the degree of inhibition of therespiratory burst. As the concentration of ECP in salmonids withBKD can vary considerably (up to 1.0 mg/ml), this relationshipcould have a significant bearing upon BKD pathogenesis. There issome precedence for dose-dependency in the immunomodulatory effectsof the ECP in that Kaattari et al. (1989) reported ECP inducedactivation of large, esterases-positive, vacuolated cells(presumably macrophages) in a dose-dependent manner. As results donot statistically confirm a dose-dependent relationship in thisinvestigation, further investigation is warranted to confirm ordispute it. Clarification of other aspects of the ECP-phagocyteinteraction in R. salmoninarum infected salmonids could prove quitebeneficial to BKD control.

-114-

Table 1: Mean cellular fluorescence(%) of splenic cells in the 10samples for each of the four treatment groups.

Group A = PBS control; Group B = 0.1 mg/ml ECP; Group C = 1.0 mg/mlECP; Group D = R. salmoninarum cells.

Fish # A B C D

1 18.5 9.3 6.4 7.6

2 13.6 10.2 8.0 8.9

3 33.9 11.3 8.7 13.0

4 57.1 29.4 18.6 29.4

5 18.3 12.1 9.5 10.1

6 13.2 7.5 6.4 6.2

7 13.5 10.2 4.9 6.8

8 22.6 10.1 11.4 8.4

9 17.1 8.0 11.2 9.6

10 22.2 12.1 12.3 13.8

-115-

Table 2: Mean cellular phagocytosis (%) of microspheres for the sixsplenic cell isolates in each of four treatment groups.

Group A = PBS control; Group B = 0.1 mg/ml ECP; Group C = 1.0mg/ml ECP; Group D = R. salmoninarum cells.

Fish # A B C D

1 10.7 12.7 11.0 11.7

2 16.0 15.3 12.3 9.7

3 10.3 13.0 9.0 10.0

4 13.7 12.7 13.3 4.7

5 6.7 8.0 13.0 11.0

6 8.0 5.3 9.3 12.3

-116-

Figure 1. Phagocytosis of fluorescent microspheres by splenicphagocytic cell of a brook trout.

(ch3fig1.jpg; 507 K)

Figure 2. Mean cellular fluorescence (respiratory burst activity)of splenic phagocytes in the four treatment groups.

(ch3fig2.pic; 264 K)

Group A= PBS control; Group B= 0.1 mg/ml ECP; Group C= 1.0 mg/mlECP; Group D= R. salmoninarum cells. * denotes significantdifference from control group (Group A).

Figure 3. Mean cellular phagocytosis of fluorescent microspheresin the four treatment groups.

(ch3fig3.pic; 288 K)

Group A= PBS control; Group B= 0.1 mg/ml ECP; Group C= 1.0 mg/mlECP; Group D= R. salmoninarum cells.

-117-

References:

Anderson, D.P., 1992. Chapter 8 - Immunization of fish spleensections and NBT, phagocytosis, PFC and antibody assays formonitoring the immune response. In: J.S. Stolen, T.C. Fletcher,D.P. Anderson, S.L. Kaattari, and A.F. Rowley (Editors), Techniques in Fish Immunology, FITC-2. SOS Publications, FairHaven, NJ, USA. pp.79-87.

Bandin, I., Ellis, A.E., Barja, J.L., and Secombes, C.J., 1993a. Interaction between rainbow trout macrophages and Renibacterium salmoninarum in vitro . Fish & Shellfish Immunology, 3:25-33.

Bandin, I., Santos, Y., Bruno, D.W., Raynard, R.S., Toranzo,A.E., and Barja, J.L., 1991. Lack of biological activity in theextracellular products of Renibacterium salmoninarum . CanadianJournal of Fisheries and Aquatic Sciences, 48:421-425.

Bandin, I., Santos, Y., Rivas, C., Skarmeta, A.M., and Barja,J.L., 1993-b. Studies on the survival of Renibacteriumsalmoninarum within salmonid macrophages. In: A. Cervino, A.Landin, A. de Coo, A. Guerra, and M. Torre (Editors), Actas delIV Congreso Nacional de Acuicultura, Centro de InvestigacionesMarinas 1993, Pontevedra, Spain. pp. 575-580.

Bass, D.A., Parce, J.W., Dechatelet, L.R., Szejda, P., Seeds,M.C., and Thomas, M., 1983. Flow cytometric studies ofoxidative product formation by neutrophils: A graded response tomembrane stimulation. Journal of Immunology, 130:1910-1917.

Brattgjerd, S., Evenson, O., and Lauve, A., 1994. Effect ofinjected yeast glucan on the activity of macrophages in Atlanticsalmon, Salmo salar L., as evaluated by in vitro hydrogenperoxide production and phagocytic capacity. Immunology, 83(2):288-294.

Bullock, G.L., Griffin, B.R., and Stuckey, H.M., 1980. Detection of Corynebacterium salmonis by direct fluorescentantibody test. Canadian Journal of Fisheries and AquaticScience, 37:719-721.

Chung, S. and Secombes, C.J., 1988. Analysis of eventsoccurring within teleost macrophages during the respiratoryburst. Comparative Biochemistry and Physiology, 89B:539-544.

Daly, J.G. and Stevenson, R.M.W., 1987. Hydrophobic andhemagglutinating properties of Renibacterium salmoninarum . Journal of General Microbiology, 133:3575-3580.

-118-

Evendon, A.J., Grayson, T.H., Gilpin, M.L., and Munn, C.B., 1993. Renibacterium salmoninarum and bacterial kidney disease -the unfinished jigsaw. Annual Review of Fish Disease, 1993:87-104.

Fielder, F. and Draxl, R., 1986. Biochemical and immunochemicalproperties of the cell surface of Renibacterium salmoninarum . Journal of Bacteriology, 168(2):799-804.

Flano, E., Lopez-Fierro, P., Razquin, B., Kaattari, S.L., andVillena, A., 1996. Histopathology of the renal and splenichematopoietic tissues of coho salmon Oncorhynchus kisutchexperimentally infected with Renibacterium salmoninarum . Diseases of Aquatic Organisms, 24:107-115.

Griffiths, S.G. and Lynch, W.H., 1991. Instability of the majorsoluble antigen produced by Renibacterium salmoninarum . Journalof Fish Diseases, 14:55-66.

Hardie, L.J., Ellis, A.E., and Secombes, C.J., 1994. In vitroactivation of rainbow trout macrophages stimulates killing ofRenibacterium salmoninarum . International Symposium on AquaticAnimal Health: Program and Abstracts. 1994, p. W3.1.

Holladay, S.D., Smith, S.A., El-Habback, H., and Caceci, T., 1996. The influence of chlorpyrifos, an organophosphateinsecticide, on the immune system of tilapia ( Oreochromisniloticus ). Journal of Aquatic Animal Health, 8:104-110.

Hsu, H.M., Bowser, P.R., and Schachte, J.H. Jr., 1991. Development and evaluation of a monoclonal-antibody-based enzyme-linked immunosorbent assay for the diagnosis of Renibacteriumsalmoninarum infection. Journal of Aquatic Animal Health, 3:168-175.

Kaattari, S.L., Turaga, P., and Wiens, G., 1989. Developmentof a vaccine for bacterial kidney disease in salmon. FinalReport 1989. U.S. Department of Energy, Bonneville PowerAdministration, Division of Fish and Wildlife, Portland, OR, USA. 310 pp.

Klein, T.C., 1992. Intracellular survival and multiplication ofRenibacterium salmoninarum within cultured macrophages fromrainbow trout ( Oncorhynchus mykiss ). University of Prince EdwardIsland Volume 32/01 of Masters Abstracts, p. 181.

-119-

Klontz, G.W., 1994. Chapter 13 - Fish hematology. In: J.S.Stolen, T.C. Fletcher, A.F. Rowley, J.T. Zelikoff, S.L. Kaattari,and S.A. Smith (Editors), Techniques in Fish Immunology-3. SOSPublications, Fair Haven, NJ, USA. pp.121-130.

Rockey, D.D., Turaga, P.S.D., Wiens, G.D., Cook, B.A.,andKaattari, S.L., 1991. Serine proteinase of Renibacteriumsalmoninarumdigests a major autologous extracellular and cell surfaceprotein. Canadian Journal of Microbiology, 37:758-763.

Rose, A.S. and Levine, R.P., 1992. Complement-mediatedopsonization and phagocytosis of Renibacterium salmoninarum . Fish & Shellfish Immunology, 2(3):223-240.

Sakai, M., Atsuta, S., and Kobayashi, M., 1989. Protectiveimmune response in rainbow trout, Oncorhynchus mykiss , vaccinatedwith B-haemolytic streptococcal bacterin. Fish Pathology, 24(3):169-173.

Sakai, M., Miyama, K., Atsuta, S., and Kobayashi, H., 1996. Thechemiluminescent responses of phagocytic cells of coho salmon,rainbow trout, and carp against Renibacterium salmoninarum . Fish & Shellfish Immunology, 6:71-73.

Secombes, C.J. and Fletcher, T.C., 1992. The role of phagocytesin the protective mechanisms of fish. Annual Review of FishDiseases, 1992:53-71.

Thuvander, A., Johannisson, A., and Grawe, J., 1992. Chapter 2- Flow cytometry in fish immunology. In: J.S. Stolen, T.C.Fletcher, D.P. Anderson, S.L. Kaattari, and A.F. Rowley(Editors), Techniques in Fish Immunology, 1992. SOSPublications, Fair Haven, NJ, USA. pp.19-26.

Tizard, I., 1987. Veterinary Immunology: An Introduction 3rdEd. W.B. Saunders Co., Philadelphia, PA. 401 pp.

Turaga, P., Wiens, G., and Kaattari, S., 1987. Bacterial kidneydisease: the potential role of soluble protein antigen(s). Journal of Fish Biology, 31(Supplement A):191-194.

Wolf, K. and Quimby, M.C., 1973. Fish viruses: buffers andmethods for plaquing eight agents under normal atmosphere. Applied Microbiology, 25:659-664.

-120-

Young, C.L. and Chapman, G.B., 1978. Ultrastructural aspects ofthe causative agent and renal histopathology of bacterial kidneydisease in brook trout ( Salvelinus fontinalis ). Journal of theFisheries Research Board of Canada, 35:1234-1248.

Acknowledgments:

I wish to thank my co-investigators for this study, Dr. StephenSmith and Dr. Stephen Holladay. Additionally, we wish to thank thefollowing for their contributions to this investigation: the staffof the National Fish Health Research Laboratory with particularregard to Dr. Cliff Starliper and Cindy Baker; Dr. Larisa Ford;Jeremy Boone and Joan Kalnitsky of the Virginia-Maryland RegionalCollege of Veterinary Medicine.

-121-

Chapter 4

BKD-induced renal lesions of brook trout ( Salvelinus fontinalis ) exposed via

ingestion of infected tissues

-122-

Abstract

Brook trout ( Salvelinus fontinalis ) exposed to Renibacteriumsalmoninarum via ingestion of infected tissues were evaluated forhistological evidence of renal lesions. The fish were fed R.salmoninarum -infected chinook salmon tissues incorporated into thefeed at a dose of 1.69 x 10 bacterial cells/fish. Renal tissues11

from five exposed trout were evaluated at 1, 4, 7, and 10 weekspost-exposure while unexposed trout were sampled at 0 and 4 weeks.Enzyme-linked immunosorbent assay results were positive for R.salmoninarum in renal tissues at the 4, 7, and 10 week sampleintervals for exposed fish, and no unexposed fish tested positive.A variety of renal glomerular, tubular, and interstitial parameterswere evaluated microscopically. Histological lesions wereidentified at 7 and 10 weeks post-exposure and includedfibroepithelial crescent formation, glomerular sclerosis, dilationof Bowman’s space, glomerular basement membrane thickening,mesangial matrix proliferation, tubular vacuolization and necrosis,inflammatory cell infiltration of the interstitium, fibrousconnective tissue deposition, and granuloma formation. Renaltissue evaluated via immunofluorescent technique with fluorochrome-labeled anti-trout immunoglobulin showed no evidence ofimmunoglobulin deposition along glomerular or tubular basementmembrane. These findings indicate that significant renal lesionsdeveloped in the R. salmoninarum -infected fish exposed by thegastrointestinal route within 7-10 weeks post-exposure. While theresults bear similarities to lesions reported in association withother routes of exposure and are consistent with an immune-mediatedcomponent, immunofluorescence did not confirm any immunologicalhypersensitivity.

-123-

I. IntroductionRenibacterium salmoninarum is the etiological agent of

bacterial kidney disease (BKD), a subacute to chronic,multisystemic, granulomatous disorder of salmonids. Renibacteriumsalmoninarum is a facultatively intracellular pathogen with anaffinity for the renal tissue of susceptible salmonid hosts.Natural transmission of BKD is largely vertical (Evelyn et al.,1984), although horizontal transmission via direct contact betweeninfected and uninfected salmonids and through consumption ofinfected tissues or feces also occurs (Murray et al., 1992; Balfryet al., 1996; Austin and Rayment, 1985). Some investigators(Elliott et al., 1994; Mitchum and Sherman, 1981) have postulatedthat vectors may play a role in the horizontal transmission of BKD.Experimental infection with R. salmoninarum in salmonids has beenaccomplished by a number of routes including immersion baths, feedpreparations containing viable organisms, and cohabitation withinfected fish (Wood and Wallis, 1955; Murray et al., 1992).However, the most common and reliable method of experimentalchallenge is parenteral injection of R. salmoninarum .

The kidney is a primary target organ in BKD pathogenesis.Grossly, R. salmoninarum -infected salmonid kidneys may appearuniformly pale and swollen or have focal abscesses (Fryer andSanders, 1981). Histologically, lesions characteristically containintracellular and extracellular R. salmoninarum , diffusegranulomas, necrosis of glomeruli and tubules, interstitialinfiltration with immunocytes, and interstitial edema (Bruno,1986).

Renal ultrastructure in BKD, described by Young and Chapman(1978), confirms significant glomerular and tubular damage. Recently, Sami et al. (1992), investigating the renalimmunopathogenesis associated with BKD in fish infectedintraperitoneally with R. salmoninarum , described a chronicmembranous glomerulonephritis and identified putativeimmunoglobulin deposits along the glomerular basement membrane viaboth electron microscopy and indirect immunofluorescence. Thisappeared to indicate that this glomerular lesion was associatedwith immune complex-mediated inflammation. Previous investigations(Young and Chapman, 1978) had also noted similarities between BKDrenal lesions and the pathogenesis of immune-mediatedglomerulonephritis in mammals, including subendothelial densedeposits resembling immune complexes, mesangial proliferation,mesangial dense deposits, and basement membrane thickening andirregularity. Further corroboration of an immune complex-mediatedcomponent to BKD was provided by Kaattari et al. (1989) with theidentification of soluble immune complexes in the serum of R.salmoninarum -infected salmonids.

-124-

In mammals, the pathogenesis of immune-complex glomerulopathyhas been extensively studied. Immune complex-mediatedglomerulonephritis, characterized as a Type III immunologicalhypersensitivity reaction, involves the deposition of circulating,soluble immune complexes in the glomerular filtration barrier.Complex deposition is followed by glomerular inflammation and renaltissue damage, and is mediated by both complement and cellularphagocytosis mechanisms (Tizard, 1987). The initial formation ofsoluble immune complexes is dependent upon a number of factorsrelated to antigen presentation. Soluble immune complexes generallyform in circumstances of significant antigen or antibody excess.Therefore, it is presumable that factors which determine the degreeof antigen exposure and/or antibody production will affect thelikelihood of soluble complex formation (Kumar et al., 1992).Furthermore, the route of antigen presentation may ultimatelyinfluence the formation of soluble immune complexes. Paul (1984)stated that presentation of antigen via the gastrointestinal tractmay impact any subsequent immune response as hepatic passage andphagocytosis may, to some degree, filter out any available antigen.

We report here the renal lesions associated with R.salmoninarum infection via gastrointestinal exposure (i.e.-ingestion of infected tissues). As gastrointestinal transmissionof R. salmoninarum has consistently produced clinical disease incultured salmonid populations (Wood and Wallis, 1955), it followsthat this means of infection represents an alternative, albeit lesspractical, to parenteral injection for the experimental inductionof BKD.

II. Materials and MethodsSixty yearling brook trout ( Salvelinus fontinalis ) were

maintained at the National Fish Health Research Laboratory inKearneysville, WV in each of two circular 700 l tanks supplied withrunning spring water (12.5 C). Both groups were maintained on ao

diet of commercial pelleted salmonid feed (Zeigler Brothers, Inc.,Gardners, PA). Over a four day period, the experimental group wasfed the viscera of chinook salmon ( Oncorhynchus tshawytscha )fingerlings infected with R. salmoninarum while the control groupwas fed only the pelleted ration. Abdominal viscera (liver,spleen, intestine, and kidney tissues) was aseptically removed fromthe dead fingerlings, finely macerated, and homogenized in astomacher for three 1-minute cycles. Forty-eight grams of theviscera (6 g/feeding over 2 feedings/day) were fed to the sixtytrout in the experimental group, thoroughly mixed with theirstandard pelleted ration. The concentration of R. salmoninarum inthe infected viscera was determined via the membrane filtrationfluorescent antibody test (MFAT; Elliott and Barila, 1987), and anaverage infective dose per fish was calculated. At the sampling

-125-

intervals of 1 week, 4 weeks, 7 weeks, and 10 weeks post-exposure,five fish were removed from the experimental group and sacrificedvia a lethal dose of tricaine methanesulfonate anesthetic (MS-222;Sigma Chemical Co., St. Louis, MO). Fish were necropsiedaseptically, gross organ structure was noted, and renal(mesonephric) tissue was collected from the mid-kidney region.Five control fish were sacrificed and necropsied at the samplingintervals of 0 weeks and 4 weeks in a similar manner. Samples ofrenal tissue were placed in 10% neutral buffered formalin fixativeand also frozen at -70 C. Upon completion of all samplings, theo

frozen renal tissues were thawed and tested by the enzyme-linkedimmunosorbent assay (ELISA) using a modification of the methoddescribed by Pascho and Mulcahy (1987). Formalin-fixed renaltissue samples were paraffin-embedded and prepared for histologicalevaluation. Sectioned specimens were stained with hematoxylin &eosin (H&E) and Periodic acid-Schiff (PAS) stains (Prophet et al.,1992). Conventional light microscopy was used to examine allhistological specimens.

Renal tissue from each specimen was evaluated for thefollowing criteria:

1. Glomeruli:* mesangial cellularity- the mean number of mesangial cells

per glomerular cross section (8 glomeruli examined)* mesangial matrix- presence or absence of increased mesangial

matrix (based upon the ratio of matrix to mesangial cells,categorized as present if greater than 1:1)

* sclerosis- presence or absence of glomerular sclerosis(adherence of the glomerular capillary bed to Bowman's capsule)

* Bowman's space dilation- presence or absence of dilation ofBowman's space ("present" if the glomerulus filled less than 3/4 ofBowman's space in cross-section)

* crescent formation- presence or absence of fibroepithelialcrescents (cellular proliferation) along Bowman's capsule

* basement membrane thickening- presence or absence of anythickening or irregularities along the glomerular basement membrane

2. Tubules:* hyaline- presence or absence of eosinophilic-staining

hyaline protein in the tubular lumen* tubular dilation- presence or absence of dilation of

proximal or distal renal tubules* vacuolization- vacuoles present or absent in renal tubular

cells ("present" if vacuoles were readily apparent in tubule cellsevery few high power fields or more frequently)

* necrosis- necrosis of renal tubules present (evident everyfew high power fields or more frequently) or absent

-126-

* basement membrane thickening- presence or absence of anythickening or irregularities(i.e.-splitting, incontinuity) of thetubular basement membrane 3. Interstitium:

* tubule density- the mean number of renal tubules per highpower field (over 10 fields)

* pigmentation- presence or absence of abundant pigmentationbased on the ratio of pigmented to nonpigmented interstitium(greater than 1:4 constituted "present")

* fibrous connective tissue- the presence or absence of peri-tubular/peri-glomerular fibrous connective tissue

* granulomas- the presence or absence of granulomas

Immunofluorescence of renal tissues was evaluated by amodification of the method described by Tanimoto and Ohtsuki(1996). Briefly, paraffin embedded tissue sections on microscopeslides were deparaffinized with xylene and absolute ethanol, placedin a hot bath (95 C) of target unmasking fluid (TUF; Signeto

Laboratories, Inc., Dedham, MA) for 10 minutes, and rinsed inwater. Tissue sections were then stained for 45 minutes withfluorochrome-labeled affinity purified goat anti-troutimmunoglobulin (Kirkegaard and Perry Laboratories, Inc.,Gaithersburg, MD) at 37 C. Stained slides were again rinsed ino

water, coverslips applied, and the tissue sections evaluated usingfluorescent microscopy. Unstained tissue sections from the troutserved as negative controls while splenic tissue sections werepositive controls.

Statistical evaluation of results was performed withnonparametric analyses with p=0.05. The Kruskal-Wallis One-WayANOVA was applied to numerical data (tubule density/glomerularcellularity) while the Fisher's exact test was used for theremaining categorical data (Norman and Streiner, 1994).

III. ResultsThe concentration of R. salmoninarum inoculum in the

homogenized viscera from the fingerlings was 2.11 x 10 cells/g 11

tissue, which translated into an infective dose of approximately1.69 x 10 bacterial cells/trout. Results of the ELISA for the11

control and experimentally infected fish are summarized in Table 1.Briefly, the ELISA was positive for R. salmoninarum in the 4-weekspost-exposure group with sampled fish testing positive at bothsubsequent sampling intervals as well. No tissue from control fishor fish tested 1 week post-exposure were ELISA positive for R.salmoninarum .

-127-

Gross observations revealed little evidence of disease. Noexternal or internal lesions involving organs other than the kidneywere noted in any fish. A swollen and pale grey kidney was notedfor one specimen from the 7 week post-exposure group.

Histological observations are summarized in Table 2.Statistically significant (p<0.05) histological changes were notedin the renal tissue samples from the 7 and the 10 week post-exposure groups. Tubular vacuolization and necrosis, granulomas,and decreased tubular density were all noted at both the 7 and 10week intervals (Figures 1, 2, and 3). Interstitial fibrousconnective tissue, glomerular sclerosis, fibroepithelial crescentformation, and dilation of Bowman's space were significantlygreater at the 10 week post-exposure sample time (Figures 1, 4, and5). Mesangial matrix proliferation and glomerular basementmembrane thickening/irregularity were significantly higher in the7 week sample group (Figure 6) compared to the control group, butnot significantly higher in the 10 week sample fish. Additionally,the presence of PAS-staining debris resembling tubular basementmembrane within the cytoplasmic vacuoles of interstitial cells(macrophages) in close proximity to degenerating tubules wereobserved in both the 7 and 10 week post-exposure groups (Figure 4).Bacteria were seen occasionally in the interstitium of exposedfish, both intracellularly and extracellularly.

Minimal, nonspecific immunofluorescence was observed in therenal samples. Occasional fluorescence of interstitial cells wasnoted for all sample time intervals, inclusive of control tissuesamples. One tissue sample from the 10 week post-exposure intervaldisplayed a mild, focal granular fluorescence associated with asingle glomerulus. However, no aggregated deposits of fluorescingmaterial were seen in glomeruli of either infected or non-infectedfish.

IV. DiscussionThis investigation has shown significant renal disease in the

kidneys of R. salmoninarum -infected trout exposed by ingestion ofinfected tissues. These results are similar to findings previouslyreported in association with experimentally induced and naturallyoccurring BKD (Bruno, 1986; Young and Chapman, 1978).

Histological parameters were selected for this study basedupon their reliability as indicators of renalpathology/immunopathology in mammals as well as their knownassociation with BKD-induced renal lesions in salmonids (Bruno,1986). Tubule density correlated inversely with interstitialcellularity and reflected the degree of inflammatory cell invasionof the interstitium. Likewise, the presence of granulomas and/orfibrous connective tissue, especially as peri-tubular or peri-

-128-

glomerular deposits, indicated a chronic state of granulomatousinflammatory disease. Pigmented melanomacrophages/melanomacrophageaggregates of the renal interstitium are reported to have aprominent role in inflammatory processes among fish (Wolke, 1992).The amount of vacuolization and necrosis of renal tubules, tubuledilation, and the presence of hyaline protein in the tubular lumenare all indicative of tubular cell viability and renal tubulefunction (Kumar et al., 1992). The glomerular parameters evaluatedin this investigation served as more specific indicators ofglomerulonephritis-induced disease potentially related to anhypersensitivity reaction. Mesangial cellularity and abundance ofthe mesangial matrix are factors which often increase inassociation with mammalian proliferative glomerulonephritis(Slauson and Lewis, 1979). Additionally, shrinkage of theglomerular tuft/dilation of Bowman's space, glomerular sclerosis,and fibroepithelial crescent formation are changes reported tooccur in glomerulonephritis (Wilson et al., 1982). Basementmembrane thickening and irregularity (i.e.-splitting, focalproliferation) associated with either the glomeruli or tubules mayindicate a membrane-associated immunologic disease such as Type IIor Type III hypersensitivity reactions (Keane and Michael, 1988;Robertson et al., 1977).

Inflammatory changes in the renal interstitium were evidentboth 7 and 10 weeks post-exposure, as shown through the increasedabundance of interstitial inflammatory cells, the infiltration offibrous connective tissue, and granuloma development. Degenerativetubular changes such as the vacuolization and necrosis of tubularepithelial cells also became readily apparent during this time.Glomerular changes consistent with glomerulonephritis, includingmesangial matrix proliferation, sclerosis, crescent formation, anddilation of Bowman's space, were also seen at 7 and 10 weeks post-exposure. Basement membrane thickening was observed chronicallyin both tubules and glomeruli, from 4 weeks post-exposure onward inglomeruli and from 7 weeks post-exposure onward in tubules(although statistical significance associated with this lesion wasfound only for glomerular membrane thickening in the 7 weeksample).

The ELISA results confirmed R. salmoninarum antigen present inrenal tissue as early as 4 weeks post-exposure, the samplinginterval prior to the appearance of significant renal lesions.While all renal samples were ELISA positive at 4 weeks post-exposure, only 60% and 40% of tissue samples tested positive at 7and 10 weeks respectively. This finding is possibly related to themethod of exposure acting as a potential confounder of theinvestigation; trout were merely offered the ration containinginfected viscera and were not force-fed the inoculum, thereforeequal dosage among sample fish was not assured. Moreover, it is

-129-

possible that alternative routes of exposure (i.e.- trans-gill ortrans-epithelium) occurred to a lesser degree via liberation of R.salmoninarum from the feed into the water column. Alternatively,tube-feeding the infected ration would be more likely to haveresulted in equal exposures among fish but would also haveintroduced an artificial stress factor with a potentiallysignificant effect upon host immune response.

Moreover, for the 10 week post-exposure group, ELISA resultsdo not correspond completely with histological indicators ofdisease as three fish with pathologic changes includinggranulomas, glomerular sclerosis, and dilation of Bowman's spacedid not have positive ELISA results. Absence of detectable levelsof soluble antigen in fish having lesions consistent with BKD mightreflect either antigen production or availability (i.e.intracellular sequestration of bacteria) at this stage in diseasepathogenesis. Additionally, this inconsistency between pathologyand ELISA may support an immune-mediated basis for renalpathogenesis in BKD, indicating that unbound R. salmoninarumsoluble antigen is seemingly not a prerequisite for the developmentof lesions.

Overall, the histological observations in this investigationwere consistent with pathologic changes induced by immune-mediatedhypersensitivity. Glomerular changes resemble those encountered inmembranoproliferative glomerulonephritis of mammals. Similarlesion formation has been described in association with BKD insalmonids infected by other routes of exposure (Sami et al., 1992;Young and Chapman, 1978). Thickening and irregularity of thebasement membrane of both glomeruli and tubules, as well as theappearance of phagocytosed tubular basement membrane ininterstitial cells suggests that a Type II hypersensitivityinvolving the formation of anti-basement membrane antibody may alsobe associated with BKD; however, our investigation did not provethis conclusively. This type of hypersensitivity reaction (i.e.anti-basement membrane antibody formation) has not been reportedamong any species of fish. Our immunofluorescence results were notindicative of either a Type II (i.e. linear fluorescence along thebasement membrane) or Type III (i.e. granular fluorescence alongthe glomerular basement membrane) hypersensitivity reaction. While one focal region of granular fluorescence was noted (10 weekspost-exposure), this finding was not consistent among the infectedfish. The inability of the immunofluorescence technique applied inthis investigation to demonstrate immune complex formation/deposition in renal tissues does not necessarily rule out anyimmunological hypersensitivity component to the renal disease.

-130-

Table 1. ELISA results for control and post-exposure groups .

Sample Group # Positive ELISA/Total # Tested

Control 0/10

1 week 0/5

4 weeks 5/5

7 weeks 3/5

10 weeks 2/5

-131-

Table 2: Summary of histological observations for the control andpost-exposure groups.Mesangial cellularity and tubule density are represented by themean value observed for all fish followed by the standard deviation(parentheses). All other data are categorically summarized as thenumber of $positive # (i.e.- lesion present) fish sampled/totalnumber sampled for each group. $* # indicates a statisticallysignificant difference in the parameter for the sampling intervalcompared to the unexposed control group.Glomeruli Control 1 week 4 weeks 7 weeks 10 weeksmesangialcellularity 53(12) 45(8) 48(8) 46(11) 41(2)

mesangial matrix 4/10 0/5 3/5 5/5 * 2/5

sclerosis 2/10 0/5 0/5 2/5 5/5 *

Bowman’s space 0/10 0/5 0/5 1/5 4/5 *dilation

fibroepithelial 0/10 0/5 1/5 2/5 4/5 *crescents

basement membrane 0/10 0/5 1/4 3/4 * 1/4thickening

Tubuleshyaline protein 1/10 0/5 0/5 0/5 0/5

dilation 0/10 0/5 1/5 2/5 2/5

vacuolization 3/10 3/5 3/5 5/5 * 5/5 *

necrosis 3/10 2/5 3/5 5/5 * 5/5 *

basement membrane 0/10 0/5 0/4 2/4 1/4thickening

Interstitiumtubule density 14.2(2.6) 19.2(5.0) 13.8(0.8) 8.8(3.7) * 9.6(1.5) *

pigment 4/10 2/5 1/5 0/5 2/5

granulomas 0/10 0/5 0/5 3/5 * 3/5 *

fibrous connective 0/10 2/5 2/5 2/5 4/5 *tissue

-132-

Figure 1. Decreased tubule density/interstitial cell infiltrationin a 10 week post-exposure sample.

(ch4fig1.jpg; 933 K)Tubular vacuolization and necrosis as well as dilation of

Bowman’s space in the glomerulus are also apparent.

Figure 2. Severe tubular necrosis of two adjacent renal tubules 10weeks post-exposure with cellular debris present in tubule lumens.


Figure 3. Granuloma formation in renal tissue 7 weeks post-exposure. (ch4fig3.jpg; 1.39M)

Figure 4. Severe tubular necrosis with deposition of fibrousconnective tissue 10 weeks post-exposure.

(ch4fig4.jpg; 1.26M)Phagocytosis of tubular basement membrane is also evidenced by thepresence of basement membrane fragments inside inflammatory cellsof the interstitium adjacent to degenerative tubules.

Figure 5. Development of a fibroepithelial crescent throughinflammatory cell proliferation adjacent to Bowman’s capsule of theglomerulus (10 weeks post-exposure).


Figure 6. Glomerular basement membrane thickening and splitting 7weeks post-exposure.

(ch4fig6.jpg; 1.14 M)

-133-

References:

Austin, B. and J.N. Rayment (1985): Epizootiology ofRenibacterium salmoninarum , the causal agent of bacterial kidneydisease in salmonid fish. Journal of Fish Diseases, 8, 505-509.

Balfry, S.K., L.J. Albright, and T.P.T. Evelyn (1996): Horizontal transfer of Renibacterium salmoninarum among farmedsalmonids via the fecal-oral route. Diseases of AquaticOrganisms, 25, 63-69.

Bruno, D.W. (1986): Histopathology of bacterial kidney diseasein laboratory infected rainbow trout, Salmo gairdneri Richardson,and Atlantic salmon, Salmo salar L., with reference to naturallyinfected fish. Journal of Fish Diseases, 9, 523-537.

Elliott, D.G. and T.Y. Barila (1987): Membrane filtration -fluorescent antibody staining procedure for detecting andquantifying Renibacterium salmoninarum in coelomic fluid ofchinook salmon ( Oncorhynchus tshawytscha ). Canadian Journal ofFisheries and Aquatic Science, 44, 206-210.

Elliott, D.G., R.J. Pascho, P.. Manning, and C.L. McKibben(1994): Studies on the potential for transmission ofRenibacterium salmoninarum to chinook salmon ( Oncorhynchustshawytscha ) during code-wire tagging procedures. InternationalSymposium on Aquatic Animal Health: Program and Abstracts,Seattle, WA, USA, p.W-17.5.

Evelyn, T.P.T., J.E. Ketcheson, and L. Prosperi-Porta (1984): Further evidence for the presence of Renibacterium salmoninarumin salmonid eggs and for the failure of povidone-iodine to reducethe intra-ovum infection rate in water-hardened eggs. Journal ofFish Diseases, 7, 173-182.

Fryer, J.L. and J.E. Sanders (1981): Bacterial kidney disease ofsalmonid fish. Annual Review of Microbiology, 35, 273-298.

Kaattari, S.L., P. Turaga, and G. Wiens (1989): Development of avaccine for bacterial kidney disease in salmon. Final Report,U.S. Department of Energy, Bonneville Power Administration,Division of Fish and Wildlife, Portland, OR, USA, 310 p.

Keane, W.F. and A.L. Micheal (1988): Renal Diseases, In"Immunological Diseases, 4th Ed." (ed. by M. Samler, D.W.Talmage, M.M. Frank, K.F. Austen, and H.N. Claman). Little,Brown, and Co., Boston, MA, USA, pp. 1809-1850.

-134-

Kumar, V., R.S. Cotran, and S.L. Robbins (1992): BasicPathology, 5th Ed. W.B. Saunders Company, Philadelphia, PA, USA, 772 p.

Mitchum, D.L. and L.E. Sherman (1981): Transmission of bacterialkidney disease from wild to stocked hatchery trout. CanadianJournal of Fisheries and Aquatic Science, 38, 547-551.

Murray, C.B., T.P.T. Evelyn, T.D. Beacham, L.W. Barner, J.E.Ketcheson, and L. Prosperi-Porta (1992): Experimental inductionof bacterial kidney disease in chinook salmon by immersion andcohabitation challenges. Diseases of Aquatic Organisms, 12, 91-96.

Norman, G.R. and D.L. Streiner (1994): Biostatistics, The BareEssentials. Mosby-Year Book, Inc., St. Louis, MO, USA, 260 p.

Pascho, R.J. and D. Mulcahy (1987): Enzyme-linked immunosorbentassay for a soluble antigen of Renibacterium salmoninarum , thecausative agent of salmonid bacterial kidney disease. CanadianJournal of Fisheries and Aquatic Science, 44, 183-191.

Paul, W.E. (1984): Fundamental Immunology. Raven Press, NewYork, NY, USA, 809 p.

Prophet, E.B., B. Mills, J.B. Arrington, and L.H. Serbin (1992): Laboratory Methods in Histotechnology. Armed Forces Institute ofPathology, American Registry of Pathology, Washington, D.C., USA,279 p.

Robertson, J.L., G.S. Hill, and D.T. Rowlands (1977): Tubulointerstitial nephritis and glomerulonephritis in Brown-Norway rats immunized with heterologous glomerular basementmembrane. American Journal of Pathology, 88, 53-68.

Sami, S., T. Fischer-Scherl, R.W. Hoffman, and C. Pfeil-Putzien(1992): immune complex-mediated glomerulonephritis associatedwith bacterial kidney disease in the rainbow trout ( Oncorhynchusmykiss ). Veterinary Pathology, 29, 169-174.

Slauson, D.O. and R.M. Lewis (1979): Comparative pathology ofglomerulonephritis in animals. Veterinary Pathology, 16, 135-164.

Tanimoto, T. and Y. Ohtsuki (1996): Evaluation of antibodiesreactive with porcine lymphocytes and lymphoma cells in formalin-fixed, paraffin embedded, antigen retrieved tissue sections. American Journal of Veterinary Research, 57, 853-859.

-135-

Tizard, I. (1987): Veterinary Immunology: An Introduction, 3rdEd. W.B. Saunders Company, Philadelphia, PA, USA, 401 p.

Wilson, C.B., A. Fornaseiri, P. Moullier, W. Tang, and D.M. Ward(1982): Renal Diseases, In "Basic and Clinical Immunology" (ed.by D.P. Stites, J.D. Stobo, H.H. Fudenberg, and J.V. Wells).Lange Medical Publications, Los Altos, CA, USA, pp. 557-575.

Wolke, R.E. (1992): Piscine macrophage aggregates: a review. Annual Review of Fish Diseases, 2, 91-108.

Wood, J.W. and J. Wallis (1955): Kidney disease in adult chinooksalmon and its transmission by feeding to young chinook salmon. Research Briefs, Fish Commission of Oregon, 6, 32-40.

Young, C.L. and G.B. Chapman (1978): Ultrastructural aspects ofthe causative agent and renal histopathology of bacterial kidneydisease in brook trout ( Salvelinus fontinalis ). Journal of theFisheries Research Board of Canada, 35, 1234-1248.

Acknowledgments:

I thank my co-investigators on this study, Dr. JohnRobertson, Dr. Stephen Smith, and Dr. Larisa Ford. Additionally,we wish to thank the following for their contributions to thisinvestigation:

the staff of the National Fish Health Research Laboratory,Kearneysville, WV with particular regard to Cliff Starliper,Rocco Cipriano, Darlene Bowling, and Cindy Baker;

Traci Sachs and Jill Songer of the histology laboratory atthe VMRCVM for their assistance with the preparation of thehistological specimens.

-136-

General Summary

As alluded to by a number of previous investigators, resultsof this dissertation suggest that the immunological aspects ofbacterial kidney disease (BKD) appear to be quite complex andmultifaceted. Involvement of immunomodulatory effects related tothe extracellular protein (ECP) are evidenced by these chapters.In vivo alterations in leukocyte numbers in peripheral bloodrelated to ECP exposure and challenge with Y. ruckeri included anincrease in peripherally circulating small lymphocytes andinhibition of an increase in monocytes post-challenge.Interpretation of these findings is somewhat difficult, givencurrent limitations in the understanding of fish (salmonid)hematology. It would seem logical that elevation of lymphocytenumbers represents immunostimulation while the lack of change inthe monocyte population may constitute an immunosuppressive change.Still, it is uncertain whether these changes represent deviationsfrom the "normal" ranges for hematological parameters (i.e.represent a true lymphocytosis and/or impediment of monocytosis).These findings, in conjunction with the in vitro ECP-relatedsuppression of phagocyte respiratory burst activity, suggest animmunomodulatory effect upon cell-mediated immunity. As describedin these investigations, cell-mediated immunomodulation does notappear to alter phagocytosis capabilities as no evidence of ECP-related effect upon this parameter was noted. The extent to whichany immunomodulation involves humoral immunity (and associatedlymphocytes) is uncertain as no differentiation of Ig+ and Ig-lymphocytes was performed. Otherwise, no evidence of humoralimmunomodulation, as indicated via the microagglutination titerdata, was apparent.

Exogenous facilitation of BKD development related toenvironmental variables was negligible in these results. Thisfinding may be an indication that vertical transmission of R.salmoninarum lends itself to the development of immunologicaltolerance and thereby minimizes the effects of exogenous parameterswhich influence immunocompetence. Alternatively, the results maybe a factor of an experimental design which failed to reflect theimpact of environmental stress. If the former is truly the case,this notion has serious implications for BKD control and thenecessity for continued emphasis on management techniques whichprevent or minimize vertical transmission of the pathogen.

-137-

The role of immunological hypersensitivity in BKDpathogenesis is supported by these results, although there remainssome question as to the type or types of hypersensitivity involved.Immune-complex mediated hypersensitivity (Type III) is consistentwith many of the findings related to the observedglomerulopathology in this and previous investigations. Granulomatous cell-mediated hypersensitivity (Type IV) is alsoconsistent with the BKD pathology described here and elsewhere,involving intracellular localization of antigen within macrophagesand subsequent inflammatory responses characterized by macrophageand lymphocyte infiltration, collagen deposition, and granulomaformation with focal necrosis. Moreover, results of theinvestigation suggest the possibility of an antibody-mediatedhypersensitivity (Type II) involving a specific humoral responseagainst an autologous tissue, namely, tubular basement membrane. Tubular necrosis with subsequent exposure of the basement membranecoinciding with what appears to be phagocytosed tubular basementmembrane in interstitial macrophages are consistent with theautologous tissue destruction encountered in Type IIhypersensitivity. Given the nature of findings in BKD-relatedliterature, it is likely that Type III and IV hypersensitivitiesare involved in BKD pathogenesis; further investigation iswarranted to confirm the involvement of a Type II hypersensitivityand the consistency of Type III hypersensitivity.

These characterizations of the salmonid immune response to BKDhave significant bearing upon BKD control. For instance, thelikelihood of an immune-complex mediated hypersensivity and thepotentially immunosuppressive effects of the R. salmoninarum ECPsuggest that antigenic stimulation (particularly ECP-relatedantigen) of humoral immunity in vaccine development would bedifficult to develop. Previous literature again suggests thatcell-mediated immunity presents a more favorable approach for BKDcontrol. Indeed, other pathogens such as Mycobacterium spp.,Brucella spp., and Listeria monocytogenes which share similarcharacteristics related to pathogenesis (i.e. facultativelyintracellular location, granulomatous disease involving Type IVhypersensitivity) have been controlled through the stimulation ofcell-mediated immunity. Specifically, the development of vaccineshas involved the use of live, attenuated organisms. Certainly,this approach invites a great deal of caution when applied to BKD,as the potential for the vaccine strain to regain virulence andsubsequently disseminate in an aquatic environment must beconsidered.

-138-

There are a number of avenues for future research regardingthe immunology of BKD which seem worthy of pursuit. The potentialfor development of a vaccine which would stimulate cell-mediatedimmunity, possibly incorporating a live, avirulent strain of R.salmoninarum , warrants further investigation. Other considerationsinclude: the immunomodulatory role of the ECP with reference to itsspecific antigenic constituents (p57 and others); immunomodulatoryeffects directed against parameters of cell-mediated immunity,particularly against the monocyte cell line; differences in theimmune response depending upon route of infection (i.e. verticalversus horizontal transmission of the pathogen); and differences inthe immunological responses among species and strains of salmonidswith different levels of susceptibility to BKD.

-139-

Vita

Christine Lynn Densmore was born May 2nd, 1964 in Frostburg,Maryland. She grew up in Denton, Maryland on the Delmarvapeninsula where her parents, Jack and Linda Densmore, currentlyreside.

Christine earned a B.S. degree with a major in biology andminor in history from Gettysburg College in 1986. In 1990, shereceived the D.V.M. degree from the Virginia-Maryland RegionalCollege of Veterinary Medicine. Following three years as aveterinary associate in small animal practice, Christine returnedto the VMRCVM at Virginia Tech to pursue a Ph.D. in veterinarymedical science with an emphasis upon aquatic animal medicine.Through a Cooperative Education Agreement, Christine has beenfortunate to spend a significant portion of her time as a graduatestudent at the National Fish Health Research Laboratory inKearneysville, West Virginia working on projects concerningdiseases of cultured and feral fish.

Apart from the laboratory, Christine enjoys outdoor activitiessuch as baseball and a variety of water sports, researching themedical history of the Civil War, and spending time with her eightmonth old son, Casey.

BACTERIAL KIDNEY DISEASE AND ITS EFFECT ON THE … · repeated efforts in helping me to prepare...

Documents

Transcript of BACTERIAL KIDNEY DISEASE AND ITS EFFECT ON THE … · repeated efforts in helping me to prepare...